Radar-based aerial object identification methods, devices, equipment, and media

Abstract

This invention relates to the field of data analysis technology, and discloses a method, apparatus, device, and medium for identifying aerial objects based on radar detection. The method includes: using radar equipment to detect the position and speed data of aerial objects within a target area during a detection time interval; calculating the continuous flight time of the aerial object based on the position and speed data; determining the aerial object as a suspected target flying object when the continuous flight time exceeds a flight time threshold and the speed data exceeds a flight speed threshold; acquiring multiple consecutive image frames of the suspected target flying object using photoelectric detection equipment, and extracting image feature data of the suspected target flying object from each image frame; determining whether the suspected target flying object is a target of a preset type based on the image feature data; and sending a preset form of flight warning information to a preset terminal if the suspected target flying object is a target of a preset type. This invention can improve the accuracy of aerial object identification.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a method, apparatus, device, and medium for identifying aerial objects based on radar detection. Background Technology

[0002] With the rapid development of aviation technology, the number and types of objects in the air are increasing, and air activities are becoming more frequent. While this trend brings many conveniences and development opportunities, it also poses a severe challenge to the field of air safety.

[0003] Currently, common methods for identifying and monitoring aerial objects have many limitations. Traditional single-sensor monitoring methods, such as relying solely on radar equipment, can obtain basic detection data of aerial objects to a certain extent, but they are difficult to accurately obtain detailed information such as the specific shape and characteristics of objects. As a result, when facing complex and ever-changing aerial situations, it is difficult to accurately determine the nature and potential threat level of objects based solely on radar data, which can easily lead to misjudgments or missed detections.

[0004] Therefore, improving the accuracy of aerial object recognition has become an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and medium for identifying aerial objects based on radar detection, with the main purpose of solving the problem of low accuracy in identifying aerial objects.

[0006] In a first aspect, to achieve the above objectives, the present invention provides a radar-based aerial object identification method, comprising: The system uses pre-set radar equipment to detect the position and speed of airborne objects within the target area during the detection time interval. The continuous flight time of the airborne object is calculated based on the location data and the movement speed data; When the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold, the aerial object is determined to be a suspected target flying object. Multiple consecutive image frames of the target suspicious flying object are acquired using a preset photoelectric detection device, and image feature data of the target suspicious flying object are extracted from each image frame; Based on the image feature data, determine whether the suspected flying object is a target of a preset type; If the target suspicious flying object is a target of the preset type, then a preset form of flight warning information is sent to the preset terminal.

[0007] Secondly, the present invention also provides an aerial object identification device based on radar detection, comprising: The radar data detection module is used to detect the position and speed data of airborne objects within the target area during the detection time interval using preset radar equipment. A flight time calculation module is used to calculate the continuous flight time of the airborne object based on the position data and the movement speed data; The suspicious object identification module is used to determine that the airborne object is a target suspicious flying object when the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold. The image feature extraction module is used to acquire multiple consecutive image frames of the target suspicious flying object using a preset photoelectric detection device, and extract the image feature data of the target suspicious flying object in each image frame; The target object determination module is used to determine whether the suspected flying object is a preset type of target object based on the image feature data. The flight information sending module is used to send a preset form of flight warning information to a preset terminal if the target suspicious flying object is a target of the preset type.

[0008] Thirdly, the present invention also provides an electronic device, the electronic device comprising: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the radar-based aerial object identification method described above.

[0009] Fourthly, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the radar-based aerial object identification method described above.

[0010] In this embodiment of the invention, radar equipment is used to detect the real-time position and velocity of aerial targets, improving the ability to continuously track and identify targets in complex environments and enhancing the real-time performance and accuracy of radar detection. By aggregating and calculating the original, high-dimensional temporal position and velocity data into continuous flight time, the amount of computational data is greatly reduced, and the accuracy of subsequent aerial object identification is improved. By using preset dual thresholds to perform rapid logical judgment on aerial objects initially detected by radar, early filtering of a large number of non-threat, conventional targets (such as birds and brief interference signals) can be achieved at the forefront of data stream processing. High-definition continuous image frames of suspicious flying objects are acquired through optoelectronic equipment and processed into structured image feature data, realizing the digital characterization of static and dynamic visual attributes such as the appearance, texture, and shape of targets, significantly improving the robustness of detection and identification in complex backgrounds (such as low-altitude, slow-moving, and small targets). By automatically classifying target types through image feature data and sending structured confirmation or warning information to designated terminals according to different classification results, the level of automated warning and response speed based on aerial object identification results is improved. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a schematic diagram of an application environment for an aerial object recognition method based on radar detection, according to one embodiment of the present invention. Figure 2 This is a flowchart illustrating a radar-based aerial object identification method according to an embodiment of the present invention. Figure 3 This is a schematic diagram of a process for detecting the position and speed data of an aerial object within a target area during a detection time interval using a preset radar device, according to an embodiment of the present invention. Figure 4 A functional block diagram of an aerial object identification device based on radar detection provided in an embodiment of the present invention; Figure 5 A schematic diagram of the structure of an electronic device for implementing a radar-based aerial object recognition method according to an embodiment of the present invention; Figure 6 This is another structural schematic diagram of an electronic device for implementing a radar-based aerial object recognition method according to an embodiment of the present invention.

[0013] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0014] To enable those skilled in the art to better understand the technical solutions of this disclosure, and to fully understand and implement the process of how this disclosure applies technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, not all embodiments. The embodiments of this disclosure and the various features within them can be combined with each other without conflict, and the resulting technical solutions are all within the protection scope of this disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort should fall within the protection scope of this disclosure.

[0015] It should be noted that the terms "first," "second," etc., used in this disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0016] This application provides a radar-based aerial object recognition method. The executing entity of this radar-based aerial object recognition method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the device provided in this application: a server, a terminal, etc. In other words, the radar-based aerial object recognition method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0017] This invention relates to a radar-based method for identifying aerial objects, which can be applied to applications such as... Figure 1 In this application environment, the client communicates with the server via a network. The client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The invention will now be described in detail through specific embodiments.

[0018] Reference Figure 2 The diagram shown is a flowchart illustrating an aerial object identification method based on radar detection according to an embodiment of the present invention. In this embodiment, the aerial object identification method based on radar detection includes: S1. Use preset radar equipment to detect the position and speed data of airborne objects within the target area during the detection time interval.

[0019] In this embodiment of the invention, the radar equipment is placed in a suitable position within a preset target detection area according to a certain plan and arrangement, so as to ensure that the radar can effectively perform its detection function and achieve comprehensive and accurate monitoring of air objects within the target detection area.

[0020] The radar device is an electronic device that uses electromagnetic waves to detect objects in the air. It emits electromagnetic waves into the target detection area. When the electromagnetic waves encounter an object in the air, they are reflected back. The radar device receives these reflected waves and calculates parameters such as the position, distance, and speed of the object in the air based on information such as the time delay and frequency change of the reflected waves.

[0021] Specifically, the aerial objects refer to various objects existing in the airspace where the target detection area is located, including but not limited to airplanes, helicopters, drones, birds, balloons, etc.

[0022] like Figure 3 As shown in this embodiment of the invention, the step of using a preset radar device to detect the position data and speed data of airborne objects within the target area during the detection time interval includes: Within the detection time interval, the antenna array of the radar device is controlled to periodically transmit radio frequency detection pulse signals toward the target detection area; The antenna array receives multiple echo signals reflected by airborne objects within the target detection area; The received multiple echo signals are subjected to signal enhancement processing to obtain enhanced echo signals, and the enhanced echo signals are converted into digital intermediate frequency signals; The digital intermediate frequency signal is subjected to pulse compression processing to obtain a digital intermediate frequency compressed signal, and the signal peak value of the digital intermediate frequency compressed signal is calculated. The signal peak value is used as the signal delay of the airborne object, and the radial distance of the airborne object is determined based on the signal delay. Signal analysis is performed on the corresponding echo signals of each channel of the antenna array to obtain the azimuth and elevation angles of the airborne object; The position data of the aerial object are calculated based on the radial distance, the azimuth angle, and the pitch angle. The received multiple echo signals are subjected to coherent signal processing, and the Doppler frequency shift of the coherently processed echo signals is extracted by a fast Fourier transform algorithm. The velocity data of the airborne object moving in the radar radial direction of the radar device is calculated based on the Doppler frequency shift.

[0023] In this embodiment of the invention, the position data refers to the information about the specific position of an aerial object within the target detection area obtained through position detection, which is usually represented in the form of coordinates, such as (x, y, z) coordinates in a rectangular coordinate system or distance and azimuth in a polar coordinate system; the movement speed data refers to the information about the movement speed of the aerial object obtained after movement speed detection, including the magnitude of the speed (the unit is usually meters per second, kilometers per hour, etc.).

[0024] In this embodiment of the invention, the transmitter in the radar device generates a radio frequency pulse signal. After modulation and other processing, this signal is sent to the antenna array. The control unit in the antenna array precisely controls the transmission timing and signal characteristics of each antenna element according to a preset period, so that the radio frequency pulse signal is transmitted to the target detection area with a specific beam shape and direction. The periodic transmission method can ensure continuous monitoring of the target detection area. By continuously transmitting pulse signals and receiving echoes, target information can be obtained in real time.

[0025] In detail, the antenna array can not only transmit signals but also receive reflected signals from the target. When the radio frequency detection pulse signal encounters an airborne object, each antenna element in the antenna array is in receiving mode. They will receive multiple echo signals reflected back from airborne objects at different locations within the target detection area. These echo signals differ in time and space because the distance from the airborne object to the radar varies, and the time it takes for the reflected signal to reach the antenna array also varies. At the same time, the position and motion state of the airborne object will also lead to differences in the intensity and phase characteristics of the reflected signal. The antenna array performs preliminary aggregation and preprocessing of these received echo signals and then transmits them to the receiver of the radar equipment.

[0026] Specifically, in the signal enhancement processing stage, the amplifier in the receiver amplifies the received weak echo signal to make its amplitude reach an appropriate range, and the amplified signal is filtered by a filter to remove high-frequency noise and interference signals such as clutter, so as to obtain a relatively pure enhanced echo signal.

[0027] In the process of converting to a digital intermediate frequency (IF) signal, the enhanced echo signal is first mixed with the local oscillator signal by a mixer to shift the signal frequency to the IF range. The IF analog signal is then converted into a digital IF signal by an analog-to-digital converter for subsequent digital signal processing.

[0028] Furthermore, in digital signal processing, based on the characteristics of the transmitted radio frequency detection pulse signal, corresponding matched filter coefficients are designed, and the digital intermediate frequency signal is input into the matched filter. Pulse compression processing is achieved through convolution operation. After pulse compression, the signal energy is concentrated in a narrow time range, forming a digital intermediate frequency compressed signal. By searching for the maximum value in the digital intermediate frequency compressed signal, the signal peak value can be obtained. This peak value corresponds to the strongest part of the signal reflected by the airborne object.

[0029] Specifically, since the signal peak corresponds to the time when the signal reflected from an airborne object reaches the radar, this time point is compared with the time point when the signal is transmitted to obtain the signal delay. Since the speed of electromagnetic wave propagation in air is approximately the speed of light, the radial distance of the airborne object can be determined by multiplying the signal delay by the speed of light and then dividing by two (because the distance the signal travels from the radar to the target and back to the radar is twice the distance between the target and the radar).

[0030] Furthermore, for beamforming, by adjusting the phase and amplitude weighting coefficients of the signals in each channel of the antenna array, the beam of the antenna array can be scanned in space. When the beam points to the target direction, the received target signal is the strongest. In terms of angle estimation, taking the phase comparison method as an example, by comparing the phase difference of the echo signals received by different channels of the antenna array, and combining the geometry of the antenna array and the propagation characteristics of electromagnetic waves, the azimuth and elevation angles of objects in the air can be calculated.

[0031] For example, in a uniform linear antenna array, the phase difference of the signals received by adjacent antenna elements is related to the azimuth angle of an object in the air. By measuring and analyzing these phase differences, the azimuth angle of the object in the air can be estimated.

[0032] In detail, radial distance represents the distance between an airborne object and the radar in a straight line, azimuth represents the angle of the airborne object in the horizontal plane relative to the radar's true north direction, and elevation represents the angle of the airborne object in the vertical plane relative to the horizontal plane. Based on trigonometric relationships, the x, y, and z coordinates of the airborne object in a rectangular coordinate system can be calculated using radial distance, azimuth, and elevation, thus determining the position data of the airborne object. For example, the x-coordinate is equal to the radial distance multiplied by the cosine of the azimuth and then by the cosine of the elevation; the y-coordinate is equal to the radial distance multiplied by the sine of the azimuth and then by the cosine of the elevation; and the z-coordinate is equal to the radial distance multiplied by the sine of the elevation.

[0033] Furthermore, in the coherent signal processing stage, the received multiple echo signals are phase-aligned and superimposed, which enhances the signal energy from the same airborne object, while the energy of noise and interference signals is relatively dispersed. The coherently processed echo signals are subjected to a fast Fourier transform to convert the time-domain signal into a frequency-domain signal. In the frequency domain, due to the relative motion between the airborne object and the radar equipment, the echo signal will generate a Doppler frequency shift. By searching for the peak frequency in the frequency domain signal, the Doppler frequency shift of the echo signal can be extracted.

[0034] Specifically, given the propagation speed of electromagnetic waves and the extracted Doppler frequency shift, according to the Doppler effect formula, the speed of an airborne object moving in the radar radial direction is equal to the Doppler frequency shift multiplied by the electromagnetic wave wavelength and then divided by two. The Doppler effect formula can be used to convert frequency changes into speed information, thereby obtaining the speed data of the airborne object moving in the radar radial direction of the radar equipment.

[0035] In this embodiment of the invention, the radar equipment is used to detect the position and velocity of aerial targets in real time, which improves the ability to continuously track and identify targets in complex environments, provides a reliable data foundation for the subsequent identification of aerial objects, and enhances the real-time performance and accuracy of radar detection.

[0036] S2. Calculate the continuous flight time of the airborne object based on the location data and the movement speed data.

[0037] In this embodiment of the invention, the continuous flight time refers to the total duration of continuous flight of an airborne object from the moment it enters the target detection area until the current moment or the moment it leaves the target detection area.

[0038] In this embodiment of the invention, calculating the continuous flight time of the airborne object based on the location data and the movement speed data includes: Multiple position points of the airborne object are obtained from the position data, and a first position point that meets a preset starting condition is extracted as the flight starting point of the airborne object, and a second position point that meets a preset ending condition is extracted as the flight ending point of the airborne object. Obtain the starting position data and ending position data corresponding to the flight start point and the flight end point, and calculate the flight distance of the air object based on the starting position data and the ending position data; Based on the movement speed data, a first movement speed corresponding to the flight start point and a second movement speed corresponding to the flight end point are determined. The sustained flight time of the airborne object is calculated based on the flight distance, the first moving speed, and the second moving speed.

[0039] In this embodiment of the invention, multiple position points of a certain aerial object are extracted from the position data. These position points are usually stored in the data recording system of the radar equipment in a specific coordinate form (such as x, y, z coordinates in a rectangular coordinate system). The multiple position points are filtered according to preset starting conditions, which include the motion state and position characteristics of the object. For example, the starting conditions require the aerial object to be within a certain altitude range and have a certain speed. Starting from the beginning of the flight path, each position point is checked in turn to see if it meets these conditions. Once the first position point that meets the preset starting conditions is found, it is determined as the flight starting point of the aerial object. The flight starting point marks that the aerial object has officially entered the effective detection range of the radar and started its flight process. The confirmation process of the flight termination point is similar to the above-mentioned process of the flight starting point, and will not be described in detail here.

[0040] In detail, the starting and ending position data corresponding to the flight start and end points are obtained. In a Cartesian coordinate system, the Euclidean distance formula can be used to calculate the distance between two points, that is, by calculating the square root of the sum of the squares of the differences between the coordinates of the two points on the horizontal and vertical axes.

[0041] In practical applications, this calculation process can be implemented using the mathematical function library in a programming language. For example, in Python, the sqrt function in the math library can be used to calculate the square root, thereby obtaining the flight distance.

[0042] Specifically, in order to determine the movement speed corresponding to the flight start point and the end point, it is necessary to associate the movement speed data with the start position data and the end position data. This association can be achieved through timestamps, that is, finding the movement speed data record that is closest to the time point of the flight start point and the end point. After determining the time point associated with the start position and the end position, the speed value of the corresponding time point is extracted from the movement speed data set. These speed values ​​are the first movement speed corresponding to the flight start point and the second movement speed corresponding to the flight end point.

[0043] Furthermore, the continuous flight time of an aerial object is calculated using the acceleration-time-distance calculation formula based on the flight distance, the first velocity, and the second velocity.

[0044] In this embodiment of the invention, by aggregating and calculating the original, high-dimensional temporal position and velocity data into continuous flight time, the amount of data is greatly compressed, reducing storage and transmission overhead. On the other hand, these features are statistical quantities based on time windows, which effectively smooth out instantaneous noise and measurement jitter in the original data, forming a more stable and representative target behavior descriptor, thereby improving the accuracy of subsequent aerial object recognition.

[0045] S3. Determine whether the continuous flight time is greater than a preset flight time threshold and whether the moving speed data is greater than a preset flight speed threshold.

[0046] In this embodiment of the invention, after obtaining the continuous flight time of an airborne object, it is compared with a preset flight time threshold. The preset flight time threshold is set according to the actual application scenario and needs. For example, in radar monitoring around an airport, the preset flight time threshold is set to 5 minutes to determine whether an airborne object may pose a potential threat to airport safety.

[0047] Similarly, after obtaining the speed data of an object in the air, it is compared with a preset flight speed threshold. The preset flight speed threshold is also set according to the specific application scenario. For example, in a military defense radar system, the preset flight speed threshold is set to 300 meters per second to identify suspicious targets flying at high speed.

[0048] When the continuous flight time is less than or equal to a preset flight time threshold, or when the moving speed data is less than or equal to a preset flight speed threshold, then execute S4 to terminate the identification of aerial objects in the target detection area.

[0049] In this embodiment of the invention, if either the condition of "continuous flight time is less than or equal to a preset flight time threshold" or "movement speed data is less than or equal to a preset flight speed threshold" is met, a termination identification command will be issued. Upon receiving the termination identification command, further collection and processing of flight data of airborne objects will cease.

[0050] When the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold, then execute S5 and determine that the airborne object is a target suspicious flying object.

[0051] In this embodiment of the invention, determining that the aerial object is a suspected target flying object includes: The flight audio of the aerial object is collected using a pre-set audio detection device; The flight audio features corresponding to the flight audio are extracted based on the pre-trained audio recognition model; Obtain the drone rotor audio features corresponding to the preset drone rotor sound, and calculate the audio feature similarity between the drone rotor audio features and the flight audio features; When the similarity of the audio features is greater than a preset audio feature similarity threshold, the flight audio feature is determined to be the drone rotor audio feature, and the aerial object corresponding to the flight audio feature is determined to be an initial suspicious flying object; When the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold, the initial suspicious flying object is determined to be the target suspicious flying object.

[0052] In this embodiment of the invention, within the target detection area, audio detection equipment is rationally deployed according to the size of the area, terrain features, and possible range of flight object activity. The audio detection equipment will be continuously in working condition, monitoring the surrounding environment's sound in real time. When an aerial object enters the target detection area, the sounds it generates during its flight, such as engine roar and rotor rotation, will be received by a microphone. The microphone converts the sound waves into electrical signals, which are then amplified and filtered by the signal conditioning circuit inside the audio detection equipment to enhance the effective signal and suppress noise interference. The processed audio signal is sampled and converted into a digital signal, which is then stored in the device's memory for subsequent analysis and processing.

[0053] In detail, the pre-trained audio recognition model is a deep learning model trained on a large number of known audio samples, such as a convolutional neural network (CNN) or a recurrent neural network (RNN). When flight audio is collected, the audio data is input into the pre-trained audio recognition model, which will perform multi-level feature extraction on the flight audio signal.

[0054] In the early layers of the audio recognition model, some basic feature extraction is performed, such as extracting time-domain features (e.g., amplitude, frequency) and frequency-domain features (e.g., spectrum, Mel-frequency cepstral coefficients). As the model deepens, more advanced and abstract features are extracted. These features can better represent the essential attributes of flight audio. Through these layers of model processing, the flight audio features corresponding to the flight audio are finally extracted. These features can reflect the unique patterns and characteristics of flight audio.

[0055] Furthermore, the drone rotor audio features corresponding to preset drone rotor sounds are obtained from a pre-stored database. Feature similarity calculation methods are then used to compare the similarity between the drone rotor audio features and the flight audio features. Feature similarity calculation can employ various methods, such as distance-based methods that calculate the distance between two feature vectors, with closer distances indicating higher similarity; or similarity-based methods, such as cosine similarity, which measures the similarity by calculating the cosine of the angle between two feature vectors. These calculation methods yield a value representing the similarity between the two features, reflecting the degree of matching between the flight audio features and the drone rotor audio features.

[0056] The calculated audio feature similarity is compared with a preset threshold. If the audio feature similarity is greater than the preset threshold, it means that the flight audio feature is very close to the drone rotor audio feature. The flight audio feature is then identified as the drone rotor audio feature. Based on this judgment, the aerial object corresponding to the flight audio feature is further identified as the initial suspicious flying object.

[0057] Furthermore, by combining the previously obtained continuous flight time and speed data with preset flight time and speed thresholds, the system checks whether the continuous flight time exceeds the preset flight time threshold and whether the speed data exceeds the preset flight speed threshold. Only when both conditions are met simultaneously is the initial suspicious flying object identified as the target suspicious flying object. This improves the accuracy of identifying target suspicious flying objects and ensures that only those aerial objects that possess both drone sound characteristics and have a long flight time and high flight speed are ultimately identified as target suspicious flying objects, so that appropriate handling measures can be taken.

[0058] In this embodiment of the invention, by using preset dual thresholds to perform rapid logical judgment on aerial objects initially detected by radar, early filtering of a large number of non-threat, conventional targets (such as birds and brief interference signals) can be achieved at the forefront of data stream processing, thereby improving the accuracy of aerial object identification.

[0059] S6. Use a preset photoelectric detection device to collect multiple consecutive image frames of the target suspicious flying object, and extract the image feature data of the target suspicious flying object in each image frame.

[0060] In this embodiment of the invention, the photoelectric detection device is a device that uses light signals to detect, identify, and track objects in the air based on optical and electrical principles. It can receive light signals reflected or radiated by objects in the air and obtain the shape, size, color, and motion state of the objects in the air through photoelectric conversion, signal processing, and other technologies. The image frame is the basic unit that constitutes a video or a continuous image sequence. In a video, each still image can be regarded as an image frame. Multiple consecutive image frames played in a certain time sequence form a dynamic video effect. In this invention, the photoelectric detection device (such as a camera) samples objects in the air at a certain frequency during the shooting process. Each sampled image is an image frame. For example, the common camera shooting frame rate is 25 or 30 frames per second, which means that 25 or 30 image frames are collected per second.

[0061] In detail, the image feature data is information extracted from image frames that describes the characteristics of a target object. This data can reflect features such as the object's shape, color, texture, and size. Common types include shape features: such as the outline, boundaries, and aspect ratio of an aerial object; color features: including the color distribution and color histogram of an aerial object; and texture features: describing the surface texture of an aerial object, such as roughness, smoothness, and graininess. By extracting and analyzing image feature data, information such as the type of a suspicious flying object can be determined, providing support for subsequent decision-making and processing.

[0062] In this embodiment of the invention, extracting image feature data of the suspected flying object in each image frame includes: Each of the image frames is subjected to noise reduction and enhancement processing to obtain multiple enhanced image frames; The foreground region containing the suspected target flying object is segmented in each of the enhanced image frames using a preset background subtraction method; Based on the foreground region, the image contour boundary of the target suspicious flying object is extracted, and the feature region image block of the target suspicious flying object is determined according to the image contour boundary; Calculate the morphological features, texture features, and color features of the image patch in the feature region; The morphological features, texture features, and color features are normalized and fused to generate image feature data that characterizes the target suspicious flying object.

[0063] In this embodiment of the invention, images are subject to various noise interferences during acquisition, such as Gaussian noise and salt-and-pepper noise. In spatial domain filtering, the common mean filter replaces the current pixel value with the average value of pixels in the pixel's neighborhood to smooth the image and reduce noise. For example, for a pixel and its eight neighboring pixels, the average of these nine pixel values ​​is used as the new value for that pixel. Image enhancement includes histogram equalization enhancement methods, which redistribute the grayscale values ​​of image pixels to make the histogram distribution of the image more uniform, thereby enhancing the image contrast. After noise reduction and enhancement processing, multiple enhanced image frames are obtained.

[0064] In detail, a background model is pre-established. This model can be obtained by statistically analyzing multiple image frames acquired over a period of time, such as calculating the average grayscale value at each pixel location as the background pixel value. After obtaining the background model, for each enhanced image frame, the grayscale value of each pixel in the image frame is subtracted from the grayscale value of the corresponding pixel in the background model. If the difference is greater than a preset threshold, the pixel is considered to belong to the foreground region, i.e., the area where the suspected target flying object is located. If the difference is less than or equal to the threshold, the pixel is considered to belong to the background region. In this way, the foreground region containing the suspected target flying object in the enhanced image frame can be segmented from the background.

[0065] Specifically, commonly used edge detection operators include the Sobel operator and the Canny operator. The Sobel operator detects edges by calculating the gradients of the image in the horizontal and vertical directions. It performs convolution operations on the image in the horizontal and vertical directions to obtain gradient values ​​in the two directions, and then determines the edges based on these two gradient values. In addition, it extracts the image contour boundary of the suspected flying object from the foreground region. Based on the extracted image contour boundary, the approximate range of the suspected flying object in the image can be determined. The largest rectangular region inside the contour or a specific region based on the shape characteristics of the target object can be selected as the feature region image block of the suspected flying object.

[0066] Furthermore, morphological processing mainly includes erosion and dilation operations. Erosion reduces the area of ​​the target object, removing small bumps and noise points from the edges. Dilation expands the area of ​​the target object, filling in small holes inside. By calculating the changes in the feature region image blocks after erosion and dilation operations, morphological features such as the target's area, perimeter, and aspect ratio can be obtained.

[0067] Common texture analysis methods include the gray-level co-occurrence matrix (GLCM) method, which calculates the gray-level combinations of pixel pairs at a certain distance and direction in an image to obtain the GLCM, from which texture features such as energy, entropy, and contrast are extracted.

[0068] Color features can be described using color histograms. A color histogram counts the frequency of different colors in an image. It divides the color space into several intervals and counts the number of pixels in each interval. For example, in the RGB color space, each color channel is divided into several levels, and the number of pixels in each level combination is counted to obtain a color histogram. By analyzing the color histogram, we can understand the color distribution of the target object and thus extract color features.

[0069] Furthermore, since morphological features, texture features, and color features may have different value ranges and dimensions, normalization is required to make them comparable in subsequent processing. A common normalization method is min-max normalization, which maps feature values ​​to the interval [0, 1]. For example, for a feature vector, the minimum and maximum values ​​are found, and each feature value is converted into a value between 0 and 1 using a formula. Through normalization, the dimensional differences between different features can be eliminated.

[0070] Feature fusion is the process of combining different types of features. It can be done by simple concatenation, which combines normalized morphological features, texture features, and color features in a certain order to form a new feature vector. Alternatively, it can be done by weighted fusion, which assigns different weights to different features based on their importance and then adds the weighted features together to obtain the fused feature vector. Through feature fusion, multiple feature information of a suspicious flying object can be combined to generate more comprehensive and representative image feature data for subsequent target recognition.

[0071] Optionally, a pre-trained target recognition AI model, such as YOLO, Faster R-CNN, or other network models, can be invoked to identify whether a target of a preset type exists in the captured consecutive image frames.

[0072] In this embodiment of the invention, high-definition continuous image frames of the target suspicious flying object are acquired by optoelectronic equipment and processed into structured "image feature data". This enables the digital characterization of the static and dynamic visual attributes of the target, such as appearance, texture, and shape, providing a core data foundation for aerial object recognition and significantly improving the robustness of detection and recognition in complex backgrounds (such as low-altitude, slow-moving, and small targets).

[0073] S7. Determine whether the suspected flying object is a target of a preset type based on the image feature data.

[0074] In this embodiment of the invention, the target objects of the preset type include various types of aircraft (such as airplanes and helicopters), drones, etc., and also include drones, birds, etc. used for agricultural operations in agricultural monitoring scenarios, so as to monitor and manage the situation in farmland.

[0075] In this embodiment of the invention, determining whether the suspected flying object is a target of a preset type based on the image feature data includes: Acquire flight plan data of the aircraft within the target detection area, and extract flight feature data of the aircraft corresponding to the flight plan of the aircraft; Calculate the feature similarity between the flight feature data and the image feature data; The feature similarity is compared with a preset feature similarity threshold, and the comparison result is used to determine whether the target suspicious flying object is a target of a preset type.

[0076] In this embodiment of the invention, the flight plan data of the flying body is collected within the target detection area. For example, it is connected with the databases of relevant institutions such as aviation management departments and drone operators. Before the flying body performs a flight mission, these institutions will require the flying body owner or operator to submit a flight plan, including information such as flight time, flight route, flight altitude, and flight speed.

[0077] Different flight plan data for different aircraft correspond to different flight characteristic data. For example, in low-altitude logistics transportation scenarios, the flight characteristic data includes a fixed flight path and direction of travel, uniform speed, and no obstacles. The flight path and direction of travel can be analyzed and judged by combining a pre-built digital twin map of the target detection area and radar detection data, which will not be elaborated here. No obstacles means that in a general flight path, if the flight altitude of the aircraft is higher than the height of the obstacle, and the difference between the flight altitude and the obstacle height is greater than a preset difference, then the flight altitude is a safe altitude, and there is no need to raise the flight altitude to avoid obstacles. It is considered to be without obstacles. Specifically, the digital twin map of the target detection area can be called, and radar detection data can be combined to analyze and judge whether there are obstacles or no obstacles during flight, but this will not be elaborated here.

[0078] In detail, after obtaining the flight plan data reported by the aircraft, it is necessary to extract data related to the flight characteristics of the aircraft. The flight characteristic data includes the model of the aircraft (different models of aircraft differ in appearance, flight performance, etc.), the expected flight trajectory (a flight route composed of a series of latitude and longitude coordinates), the flight altitude range, the flight speed range, etc.

[0079] Since flight feature data and image feature data are obtained from different sources and may have different representations and dimensions, they need to be aligned before similarity calculation. For example, the model feature of an aircraft may be in text form (such as "DJI Mavic 3 drone") in flight feature data, while it may be some parameters obtained through image recognition in image feature data. By establishing a feature mapping relationship, the text-based model feature can be converted into a form that matches the relevant parameters in the image feature data. For features such as flight trajectory, flight altitude, and flight speed, corresponding unit unification and format conversion are also required to ensure that the two types of feature data are compared on the same benchmark.

[0080] Specifically, based on feature alignment, various methods can be used to calculate the similarity between flight feature data and image feature data. A common method is based on the similarity comparison of feature attributes. For example, for the model of the aircraft, if the two match perfectly, the similarity is high; if they match partially or not at all, the similarity is low. For flight trajectories, the degree of spatial overlap can be compared; if the two flight trajectories overlap in most areas, the similarity is high. For flight altitude and flight speed, their numerical ranges can be compared; if their numerical ranges have a large overlap, the similarity is high. By comprehensively considering the similarity of various feature attributes, a total feature similarity can be calculated.

[0081] Furthermore, the calculated feature similarity is compared with a preset feature similarity threshold. If the feature similarity is greater than or equal to the preset threshold, it indicates that the features of the suspected target flying object are highly similar to the features of the flying body in the reported flight plan. It can be determined that the suspected target flying object is a target of the preset type, that is, it may be a legal flying body that has been reported and is flying according to the plan. If the feature similarity is less than the preset threshold, it indicates that the features of the suspected target flying object are significantly different from the features of the flying body in the reported flight plan. It can be determined that the suspected target flying object is not a target of the preset type and may belong to a flying body that has not been reported or has abnormal flight behavior. Further measures need to be taken to deal with it, such as issuing an alarm and conducting tracking and monitoring.

[0082] If the target suspicious flying object is not a target of the preset type, then execute S8 and send a preset type of flight confirmation information to the preset terminal.

[0083] In this embodiment of the invention, sending flight confirmation information in a preset format to a preset terminal includes: If the feature similarity is greater than or equal to a preset feature similarity threshold, then the target suspicious flying object is determined not to be a target of a preset type. When it is determined that the target suspicious flying object is not a target of the preset type, a flight confirmation message of the preset format is sent to the preset terminal, or a flight voice prompt of the preset format is broadcast to the preset terminal through a preset voice broadcasting device.

[0084] In this embodiment of the invention, when it is determined that the suspected target flying object is not a target of a preset type, a flight confirmation message in the format of "An unidentified flying object has appeared in the target detection area. Please confirm with the flight responsibility unit reported for the current time period whether the flying object is the normal flight behavior of the other party" is sent to a predetermined electronic device. Alternatively, a voice prompt of "An unidentified flying object has appeared in the target detection area. Please confirm with the flight responsibility unit reported for the current time period whether the flying object is the normal flight behavior of the other party" is broadcast through a voice broadcasting device.

[0085] In this embodiment of the invention, when the calculated feature similarity is compared with a preset feature similarity threshold, if the feature similarity is greater than or equal to the preset feature similarity threshold, it is determined that the target suspicious flying object is not a target of the preset type.

[0086] Specifically, when it is determined that the suspected flying object is not a target of a preset type, the flight confirmation information in a preset format is encapsulated. The encapsulation process includes packaging the flight confirmation information according to the requirements of the network protocol, adding necessary protocol header information (such as data length, checksum, etc.), and sending the encapsulated data to a preset terminal through the established network connection.

[0087] Optionally, a speech synthesis engine (such as Baidu Speech Synthesis API, iFlytek Speech Synthesis Engine, etc.) is used to convert the generated text into speech data. The speech synthesis engine converts the text into corresponding speech waveform data. The generated speech waveform data usually needs to be encoded to reduce the amount of data and improve transmission efficiency. Common audio encoding formats include MP3, WAV, etc. The encoding process compresses and converts the speech waveform data according to a certain algorithm to generate an audio file that conforms to the encoding format. The encoded audio data is sent to the preset terminal through a communication link (such as a wired connection or a wireless network connection) between the preset voice broadcasting device and the preset terminal, thereby realizing the voice broadcasting function.

[0088] If the target suspicious flying object is a target of the preset type, then execute S9 and send a preset form of flight warning information to the preset terminal.

[0089] In this embodiment of the invention, sending flight warning information in a preset format to a preset terminal includes: If the feature similarity is less than a preset feature similarity threshold, then the target suspicious flying object is determined to be a target object of the preset type; When it is determined that the target suspicious flying object is a target of a preset type, structured early warning data of the target suspicious flying object is generated based on the flight data of the target suspicious flying object; The structured early warning data is encapsulated into a data stream that conforms to data transmission specifications according to the preset terminal type and communication interface; Send the data stream and a flight warning message in a preset format to the terminal, or broadcast a voice warning message in a preset format to the terminal via a preset voice broadcasting device.

[0090] In this embodiment of the invention, for example, a flight warning message in the format of "An unidentified flying object has appeared in the target detection area, please pay attention" is sent to a predetermined electronic device, or a voice warning prompt of "An unidentified flying object has appeared in the target detection area, please pay attention" is broadcast through a voice broadcasting device.

[0091] In this embodiment of the invention, when the calculated feature similarity is compared with a preset feature similarity threshold, if the feature similarity is less than the preset feature similarity threshold, the target suspicious flying object is determined to be a target of a preset type.

[0092] Specifically, when it is determined that the target suspicious flying object is a target of a preset type, the flight data of the target suspicious flying object is acquired, and the extracted flight data is organized according to a predetermined structured format. For example, a structure containing multiple fields can be designed, such as "flight time field", "flight route field", "flight altitude field", "flight speed field", etc. The flight data is filled into the corresponding fields to form a structured data set, and then a structured early warning data object is generated.

[0093] The terminal types may include smartphones, tablets, computers, etc., and the communication interfaces may include Wi-Fi, Bluetooth, 4G / 5G networks, etc. Based on the identified terminal type and communication interface, an appropriate data encapsulation method is selected. For example, for smartphones connected via Wi-Fi, JSON format can be used for data encapsulation. The structured warning data is converted into a JSON format string, and necessary protocol header information (such as data type identifier, data length, etc.) is added to form a data stream that conforms to the data transmission standard.

[0094] Specifically, based on the communication interface of the preset terminal, the corresponding network connection method is selected. For example, if the communication interface is Wi-Fi, the Wi-Fi protocol will be used to establish a connection with the terminal, and the encapsulated data stream and the flight warning information in the preset format will be sent to the preset terminal. For the data stream, it will be transmitted through the established network connection according to the previously encapsulated format and protocol. For the flight warning information, it will also be packaged and sent in accordance with the requirements of the network protocol.

[0095] Optionally, a speech synthesis engine (such as Baidu Speech Synthesis API, iFlytek Speech Synthesis Engine, etc.) is used to convert text into speech data. The generated speech data is encoded to reduce the amount of data and improve transmission efficiency. The encoded audio data is sent to the preset terminal through a communication link (such as a wired connection or a wireless network connection) between the preset voice broadcasting device and the preset terminal. After receiving the audio data, the preset terminal decodes and plays it, thereby realizing the voice broadcasting function.

[0096] In this embodiment of the invention, by automatically classifying target types from image feature data, the capabilities of computer vision and pattern recognition algorithms are fully utilized. Based on different classification results, differentiated downstream task processes are triggered: structured confirmation or warning information is sent to designated terminals, transforming information transmission from raw data broadcasting to precise push based on threat level. This improves the accuracy of aerial object recognition and significantly enhances the level of automated warning and response speed based on aerial object recognition results.

[0097] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0098] like Figure 4 The diagram shown is a functional block diagram of an aerial object identification device based on radar detection provided in an embodiment of the present invention.

[0099] In this embodiment of the disclosure, an aerial object recognition device based on radar detection is provided, which corresponds one-to-one with the aerial object recognition method based on radar detection described in the above embodiments. For example... Figure 4 As shown, this radar-based aerial object identification device 100 can be installed in an electronic device. According to its functions, the radar-based aerial object identification device 100 includes a radar data detection module 101, a flight time calculation module 102, a suspicious object identification module 103, an image feature extraction module 104, a target object judgment module 105, and a flight information transmission module 106. Detailed descriptions of each functional module are as follows: The radar data detection module 101 is used to detect the position data and speed data of airborne objects within the target area during the detection time interval using a preset radar device. Flight time calculation module 102 is used to calculate the continuous flight time of the airborne object based on the position data and the movement speed data; The suspicious object identification module 103 is used to determine that the airborne object is a target suspicious flying object when the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold. The image feature extraction module 104 is used to acquire multiple consecutive image frames of the target suspicious flying object using a preset photoelectric detection device, and extract the image feature data of the target suspicious flying object in each image frame; The target object determination module 105 is used to determine whether the suspected flying object is a preset type of target object based on the image feature data. The flight information sending module 106 is used to send a preset form of flight warning information to a preset terminal if the target suspicious flying object is a target object of the preset type.

[0100] In one embodiment, when the radar data detection module 101 performs the task of detecting the position data and speed data of airborne objects within a target area during a detection time interval using a preset radar device, it is used to: Within the detection time interval, the antenna array of the radar device is controlled to periodically transmit radio frequency detection pulse signals toward the target detection area; The antenna array receives multiple echo signals reflected by airborne objects within the target detection area; The received multiple echo signals are subjected to signal enhancement processing to obtain enhanced echo signals, and the enhanced echo signals are converted into digital intermediate frequency signals; The digital intermediate frequency signal is subjected to pulse compression processing to obtain a digital intermediate frequency compressed signal, and the signal peak value of the digital intermediate frequency compressed signal is calculated. The signal peak value is used as the signal delay of the airborne object, and the radial distance of the airborne object is determined based on the signal delay. Signal analysis is performed on the corresponding echo signals of each channel of the antenna array to obtain the azimuth and elevation angles of the airborne object; The position data of the aerial object are calculated based on the radial distance, the azimuth angle, and the pitch angle. The received multiple echo signals are subjected to coherent signal processing, and the Doppler frequency shift of the coherently processed echo signals is extracted by a fast Fourier transform algorithm. The velocity data of the airborne object moving in the radar radial direction of the radar device is calculated based on the Doppler frequency shift.

[0101] In one embodiment, when the flight time calculation module 102 calculates the continuous flight time of the airborne object based on the position data and the movement speed data, it is used to: Multiple position points of the airborne object are obtained from the position data, and a first position point that meets a preset starting condition is extracted as the flight starting point of the airborne object, and a second position point that meets a preset ending condition is extracted as the flight ending point of the airborne object. Obtain the starting position data and ending position data corresponding to the flight start point and the flight end point, and calculate the flight distance of the air object based on the starting position data and the ending position data; Based on the movement speed data, a first movement speed corresponding to the flight start point and a second movement speed corresponding to the flight end point are determined. The sustained flight time of the airborne object is calculated based on the flight distance, the first moving speed, and the second moving speed.

[0102] In one embodiment, when the suspicious object identification module 103 determines that the aerial object is a target suspicious flying object, it is used to: The flight audio of the aerial object is collected using a pre-set audio detection device; The flight audio features corresponding to the flight audio are extracted based on the pre-trained audio recognition model; Obtain the drone rotor audio features corresponding to the preset drone rotor sound, and calculate the audio feature similarity between the drone rotor audio features and the flight audio features; When the similarity of the audio features is greater than a preset audio feature similarity threshold, the flight audio feature is determined to be the drone rotor audio feature, and the aerial object corresponding to the flight audio feature is determined to be an initial suspicious flying object; When the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold, the initial suspicious flying object is determined to be the target suspicious flying object.

[0103] In one embodiment, when the image feature extraction module 104 extracts image feature data of the suspected target flying object in each image frame, it is used to: Each of the image frames is subjected to noise reduction and enhancement processing to obtain multiple enhanced image frames; The foreground region containing the suspected target flying object is segmented in each of the enhanced image frames using a preset background subtraction method; Based on the foreground region, the image contour boundary of the target suspicious flying object is extracted, and the feature region image block of the target suspicious flying object is determined according to the image contour boundary; Calculate the morphological features, texture features, and color features of the image patch in the feature region; The morphological features, texture features, and color features are normalized and fused to generate image feature data that characterizes the target suspicious flying object.

[0104] In one embodiment, when the target object determination module 105 performs the task of determining whether the suspected flying object is a preset type of target object based on the image feature data, it is used to: Acquire flight plan data of the aircraft within the target detection area, and extract flight feature data of the aircraft corresponding to the flight plan of the aircraft; Calculate the feature similarity between the flight feature data and the image feature data; The feature similarity is compared with a preset feature similarity threshold, and the comparison result is used to determine whether the target suspicious flying object is a target of a preset type.

[0105] In one embodiment, when the flight information sending module 106 sends flight confirmation information in a preset format to a preset terminal, it is used to: If the feature similarity is greater than or equal to a preset feature similarity threshold, then the target suspicious flying object is determined not to be a target of a preset type. When it is determined that the target suspicious flying object is not a target of the preset type, a flight confirmation message of the preset format is sent to the preset terminal, or a flight voice prompt of the preset format is broadcast to the preset terminal through a preset voice broadcasting device.

[0106] In one embodiment, when the flight information sending module 106 sends flight warning information in a preset format to a preset terminal, it is used to: If the feature similarity is less than a preset feature similarity threshold, then the target suspicious flying object is determined to be a target object of the preset type; When it is determined that the target suspicious flying object is a target of a preset type, structured early warning data of the target suspicious flying object is generated based on the flight data of the target suspicious flying object; The structured early warning data is encapsulated into a data stream that conforms to data transmission specifications according to the preset terminal type and communication interface; Send the data stream and a flight warning message in a preset format to the terminal, or broadcast a voice warning message in a preset format to the terminal via a preset voice broadcasting device.

[0107] In this invention, the specific limitations of the radar-based aerial object identification device can be found in the above-described limitations of the radar-based aerial object identification method, and will not be repeated here. Each module in the aforementioned radar-based aerial object identification device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0108] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used for communication with external clients via a network connection. When executed by the processor, the computer program implements the functions or steps of the server-side of the radar-based aerial object recognition method.

[0109] In one embodiment, a computer device is provided, which may be a client, and its internal structure diagram may be as follows: Figure 6 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with an external server via a network connection. When executed by the processor, the computer program implements the functions or steps of the client-side of the radar-based aerial object recognition method.

[0110] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: The system uses pre-set radar equipment to detect the position and speed of airborne objects within the target area during the detection time interval. The continuous flight time of the airborne object is calculated based on the location data and the movement speed data; When the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold, the aerial object is determined to be a suspected target flying object. Multiple consecutive image frames of the target suspicious flying object are acquired using a preset photoelectric detection device, and image feature data of the target suspicious flying object are extracted from each image frame; Based on the image feature data, determine whether the suspected flying object is a target of a preset type; If the target suspicious flying object is a target of the preset type, then a preset form of flight warning information is sent to the preset terminal.

[0111] In the several embodiments provided by this invention, it should be understood that the disclosed devices and apparatuses can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0112] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0113] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0114] In some embodiments of this example, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.

[0115] The readable storage medium of the present invention stores a computer program, which, when executed by a processor of an electronic device, can perform the following: The system uses pre-set radar equipment to detect the position and speed of airborne objects within the target area during the detection time interval. The continuous flight time of the airborne object is calculated based on the location data and the movement speed data; When the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold, the aerial object is determined to be a suspected target flying object. Multiple consecutive image frames of the target suspicious flying object are acquired using a preset photoelectric detection device, and image feature data of the target suspicious flying object are extracted from each image frame; Based on the image feature data, determine whether the suspected flying object is a target of a preset type; If the target suspicious flying object is a target of the preset type, then a preset form of flight warning information is sent to the preset terminal.

[0116] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0117] Computer-readable storage media may also store at least one computer-executable program / instruction, such as computer-readable instructions. Computer-readable storage media include, but are not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Computer-readable storage media may include, for example, read-only memory (ROM), hard disk, flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device executes the computer-readable instructions stored on the computer-readable storage medium, the various methods described above can be performed.

[0118] In addition, the computer device may include (but is not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (e.g., keyboard, mouse, speakers, etc.).

[0119] In one embodiment, the at least one computer-executable instruction may also be compiled into or comprise a software product / computer program product, wherein one or more computer-executable instructions are executed by a processor to perform the steps of the various functions and / or methods in the embodiments described herein.

[0120] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Furthermore, any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory.

[0121] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0122] In the embodiments provided in this disclosure, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0123] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

[0124] It should be noted that if any software tools or components not belonging to our company appear in the embodiments of this application, they are merely for illustrative purposes and do not represent actual use.

Claims

1. A method for identifying aerial objects based on radar detection, characterized in that, The method includes: The system uses pre-set radar equipment to detect the position and speed of airborne objects within the target area during the detection time interval. The continuous flight time of the airborne object is calculated based on the location data and the movement speed data; When the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold, the aerial object is determined to be a suspected target flying object. Multiple consecutive image frames of the target suspicious flying object are acquired using a preset photoelectric detection device, and image feature data of the target suspicious flying object are extracted from each image frame; Based on the image feature data, determine whether the suspected flying object is a target of a preset type; If the target suspicious flying object is a target of the preset type, then a preset form of flight warning information is sent to the preset terminal.

2. The aerial object identification method based on radar detection as described in claim 1, characterized in that, The method of using preset radar equipment to detect the position and speed data of airborne objects within the target area during the detection time interval includes: Within the detection time interval, the antenna array of the radar device is controlled to periodically transmit radio frequency detection pulse signals toward the target detection area; The antenna array receives multiple echo signals reflected by airborne objects within the target detection area; The received multiple echo signals are subjected to signal enhancement processing to obtain enhanced echo signals, and the enhanced echo signals are converted into digital intermediate frequency signals; The digital intermediate frequency signal is subjected to pulse compression processing to obtain a digital intermediate frequency compressed signal, and the signal peak value of the digital intermediate frequency compressed signal is calculated. The signal peak value is used as the signal delay of the airborne object, and the radial distance of the airborne object is determined based on the signal delay. Signal analysis is performed on the corresponding echo signals of each channel of the antenna array to obtain the azimuth and elevation angles of the airborne object; The position data of the aerial object are calculated based on the radial distance, the azimuth angle, and the pitch angle. The received multiple echo signals are subjected to coherent signal processing, and the Doppler frequency shift of the coherently processed echo signals is extracted by a fast Fourier transform algorithm. The velocity data of the airborne object moving in the radar radial direction of the radar device is calculated based on the Doppler frequency shift.

3. The radar-based aerial object identification method as described in claim 1, characterized in that, The calculation of the sustained flight time of the airborne object based on the location data and the movement speed data includes: Multiple position points of the airborne object are obtained from the position data, and a first position point that meets a preset starting condition is extracted as the flight starting point of the airborne object, and a second position point that meets a preset ending condition is extracted as the flight ending point of the airborne object. Obtain the starting position data and ending position data corresponding to the flight start point and the flight end point, and calculate the flight distance of the air object based on the starting position data and the ending position data; Based on the movement speed data, a first movement speed corresponding to the flight start point and a second movement speed corresponding to the flight end point are determined. The sustained flight time of the airborne object is calculated based on the flight distance, the first moving speed, and the second moving speed.

4. The aerial object identification method based on radar detection as described in claim 1, characterized in that, The determination that the aerial object is a suspected target flying object includes: The flight audio of the aerial object is collected using a pre-set audio detection device; The flight audio features corresponding to the flight audio are extracted based on the pre-trained audio recognition model; Obtain the drone rotor audio features corresponding to the preset drone rotor sound, and calculate the audio feature similarity between the drone rotor audio features and the flight audio features; When the similarity of the audio features is greater than a preset audio feature similarity threshold, the flight audio feature is determined to be the drone rotor audio feature, and the aerial object corresponding to the flight audio feature is determined to be an initial suspicious flying object; When the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold, the initial suspicious flying object is determined to be the target suspicious flying object.

5. The radar-based aerial object identification method as described in claim 1, characterized in that, The step of extracting image feature data of the suspected flying object in each image frame includes: Each of the image frames is subjected to noise reduction and enhancement processing to obtain multiple enhanced image frames; The foreground region containing the suspected target flying object is segmented in each of the enhanced image frames using a preset background subtraction method; Based on the foreground region, the image contour boundary of the target suspicious flying object is extracted, and the feature region image block of the target suspicious flying object is determined according to the image contour boundary; Calculate the morphological features, texture features, and color features of the image patch in the feature region; The morphological features, texture features, and color features are normalized and fused to generate image feature data that characterizes the target suspicious flying object.

6. The radar-based aerial object identification method as described in claim 1, characterized in that, The step of determining whether the suspected flying object is a preset type of target object based on the image feature data includes: Acquire flight plan data of the aircraft within the target detection area, and extract flight feature data of the aircraft corresponding to the flight plan of the aircraft; Calculate the feature similarity between the flight feature data and the image feature data; The feature similarity is compared with a preset feature similarity threshold, and the comparison result is used to determine whether the target suspicious flying object is a target of a preset type.

7. The radar-based aerial object identification method as described in claim 1, characterized in that, Sending flight warning information in a preset format to a preset terminal includes: If the feature similarity is less than a preset feature similarity threshold, then the target suspicious flying object is determined to be a target object of the preset type; When it is determined that the target suspicious flying object is a target of a preset type, structured early warning data of the target suspicious flying object is generated based on the flight data of the target suspicious flying object; The structured early warning data is encapsulated into a data stream that conforms to data transmission specifications according to the preset terminal type and communication interface; Send the data stream and a flight warning message in a preset format to the terminal, or broadcast a voice warning message in a preset format to the terminal via a preset voice broadcasting device.

8. An aerial object identification device based on radar detection, characterized in that, The device includes: The radar data detection module is used to detect the position and speed data of airborne objects within the target area during the detection time interval using preset radar equipment. A flight time calculation module is used to calculate the continuous flight time of the airborne object based on the position data and the movement speed data; The suspicious object identification module is used to determine that the airborne object is a target suspicious flying object when the continuous flight time is greater than a preset flight time threshold and the moving speed data is greater than a preset flight speed threshold. The image feature extraction module is used to acquire multiple consecutive image frames of the target suspicious flying object using a preset photoelectric detection device, and extract the image feature data of the target suspicious flying object in each image frame; The target object determination module is used to determine whether the suspected flying object is a preset type of target object based on the image feature data. The flight information sending module is used to send a preset form of flight warning information to a preset terminal if the target suspicious flying object is a target of the preset type.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the radar-based aerial object identification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the radar-based aerial object identification method as described in any one of claims 1 to 7.

Images