Unmanned aerial vehicle information identification method, system, equipment and medium

Abstract

The invention discloses an unmanned aerial vehicle information identification method, system and device and a medium, and belongs to the field of electric power systems.The method comprises the steps that if a moving object is a black-flight unmanned aerial vehicle, when the first distance between a first imaging sensor and the black-flight unmanned aerial vehicle is smaller than or equal to the second distance between a second imaging sensor and the black-flight unmanned aerial vehicle, the black-flight unmanned aerial vehicle is identified; or when a first included angle between the first imaging sensor and the equipment identity label is smaller than or equal to a second included angle between the second imaging sensor and the equipment identity label, focusing of the equipment identity label is carried out through the first imaging sensor, and the black flight unmanned aerial vehicle is tracked through the second imaging sensor; and after the focusing is completed, the equipment identity identification is identified to obtain the identity identification information of the black flight unmanned aerial vehicle, so that complete and accurate identification of the identity information can be realized by implementing the method and the device.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) monitoring and identification, and in particular to a method, system, device and medium for UAV information identification. Background Technology

[0002] With the rapid development of drone technology, the commercial application of civilian drones is becoming increasingly widespread. However, the phenomenon of "black flights" (unauthorized flights) due to a lack of effective supervision is frequent. In the process of identifying the identity of black-flying drones, equipment identification marks (such as unique QR codes, barcodes, or serial numbers on the fuselage) are crucial. These marks record the drone's production information, registration information, and user-related data, serving as the sole credential for distinguishing an individual's identity. Black flights refer to flights conducted without a private pilot's license or with an aircraft lacking legal status—in other words, unregistered flights.

[0003] However, device identification tags are usually small (much smaller than the drone body), and their placement is strongly correlated with the drone model (e.g., most are located at the bottom, while some models are located on the side or top), requiring magnification of 2-3 times using visual equipment for clear identification.

[0004] In existing technologies, device identification relies heavily on a single imaging sensor for target tracking and identifier focusing. When the sensor focuses on a tiny device identifier, the field of view is significantly reduced, making it difficult to perceive the overall motion of the drone in real time (such as position changes, attitude adjustments, and rapid changes of direction). Conversely, prioritizing overall drone tracking prevents accurate magnification and focusing of the identifier, resulting in blurred identifier images or the drone moving out of the field of view, ultimately leading to identification failure and impacting the efficiency of tracing unauthorized drones. Especially when the distance between the drone and the sensor changes rapidly or the attitude is dynamically adjusted, a single sensor struggles to balance "focusing accuracy" and "tracking stability," easily losing the target due to the field of view compression required for focusing. This becomes a key flaw in successfully acquiring identification information from unauthorized drones. Summary of the Invention

[0005] This invention provides a method, system, device, and medium for drone information identification, which can solve the problem of drone target loss leading to identification failure.

[0006] This invention provides a method for identifying information about unmanned aerial vehicles (UAVs), comprising:

[0007] The motion image of the moving object in the monitoring area is acquired, and the motion characteristics of the moving object in the motion image are identified. If the moving object is identified as an unauthorized flying drone, the motion image is identified to obtain the drone type and attitude information of the unauthorized flying drone. Based on the drone type and attitude information, the direction information of the device identification mark of the unauthorized flying drone on the unauthorized flying drone is determined.

[0008] Calculate the first distance and first pose relationship between the first imaging sensor and the unmanned aerial vehicle (UAV) based on the motion image, and calculate the second distance and second pose relationship between the second imaging sensor and the UAV. Calculate the first angle between the first imaging sensor and the location of the device identification mark on the UAV based on the first pose relationship and the direction information, and calculate the second angle between the second imaging sensor and the location of the device identification mark on the UAV based on the second pose relationship and the direction information.

[0009] When the first distance is less than or equal to the second distance, or the first included angle is less than or equal to the second included angle, the first imaging sensor focuses on the device identification of the unauthorized drone, and the second imaging sensor tracks the unauthorized drone to assist the first imaging sensor in focusing;

[0010] Once the first imaging sensor has focused on the device identification identifier, it identifies the device identification identifier to obtain the identification information of the unauthorized drone.

[0011] This invention achieves preliminary differentiation between unauthorized drones and other moving objects (such as birds and balloons) through motion feature recognition, avoiding invalid processing. Based on the approximate direction (e.g., bottom, side) of the drone's identification device, it provides "position navigation" for subsequent precise focusing, reducing blind searching. Combining attitude information (e.g., pitch angle, roll angle) corrects the identification direction judgment, ensuring more accurate identification orientation analysis and providing a basis for focusing sensor selection. By quantifying distance (first / second distance) and angle (first / second included angle), an objective standard for focusing sensor selection is established. Sensors with closer distances (reducing focusing magnification) or smaller included angles (more complete identification field of view) focus, reducing the risk of field-of-view loss due to magnification. A second sensor tracks the overall movement, providing real-time feedback on position changes, assisting the focusing sensor in dynamically adjusting the viewing angle, resolving the contradiction of "focusing and tracking being mutually exclusive." Precise focusing achieves clear imaging of tiny identifiers (QR codes, numerical strings), ensuring complete extraction of identification information, providing a unique credential for tracing unauthorized drones, and realizing a closed loop of "detection-identification-tracing."

[0012] Furthermore, acquiring motion images of moving objects within the monitoring area specifically involves:

[0013] The system monitors the presence of the moving object within the monitoring area using radar and acquires the radar echo information. The echo information is then processed to obtain a signal processing result. Based on the signal processing result, it is determined whether a moving object exists. If so, the moving image is acquired using a first imaging sensor or a second imaging sensor.

[0014] In this way, radar 1 has the advantages of long-range and wide-area monitoring, and can detect moving objects before visual sensors, reducing the invalid scanning of visual sensors; it can make a preliminary judgment on the existence of moving objects through echo information, avoiding the waste of visual sensors in areas without targets, and improving the system response efficiency.

[0015] Furthermore, the acquisition of motion images of moving objects within the monitoring area, wherein the motion features include velocity features and motion trajectory features, specifically:

[0016] The echo information is subjected to noise filtering and clutter suppression to obtain target echo information. The target echo information is then clustered to identify various moving objects. The echo intensity of each moving object is determined based on the target echo information.

[0017] Based on the echo intensity, the motion characteristics of each moving object are identified by a convolutional neural network. If the motion trajectory characteristics of the target moving object are identified as those of a drone flight trajectory, or if the speed value of the target moving object is greater than a preset threshold, then the target moving object is determined to be the unauthorized drone, and the motion image of the target moving object is acquired.

[0018] This approach improves the purity of the echo signal through noise filtering and clutter suppression, reducing misjudgments caused by environmental interference (such as electromagnetic noise and ground clutter). Cluster analysis distinguishes multiple targets (such as the simultaneous presence of drones and birds), avoiding target overlap interference and laying the foundation for accurate identification. Echo intensity reflects the target's physical properties (such as the RCS difference between drones and birds), and combined with motion trajectories (such as hovering and fixed-point turning) and speed characteristics (drones are faster than birds), multi-dimensional joint identification is achieved through convolutional neural networks, significantly reducing the misjudgment rate (such as avoiding misidentifying high-speed birds as drones).

[0019] Furthermore, acquiring motion images of moving objects within the monitoring area specifically involves:

[0020] The echo information is subjected to noise filtering and clutter suppression to obtain target echo information. The target echo information is then clustered. If multiple moving objects are identified, the motion trajectory of each moving object is identified using a convolutional neural network. If the motion trajectory feature of the target moving object is identified as the flight trajectory feature of a UAV, the target moving object is determined to be the unauthorized UAV, and the motion image of the target moving object is acquired.

[0021] By focusing on the differences between the unique motion trajectories of drones (such as constant speed cruise, vertical take-off and landing, and minimal irregular swaying) and other moving objects (such as the random trajectories of birds), and by using deep learning models to enhance the ability to distinguish trajectory features, the accuracy of identifying unauthorized drones is further improved. This is especially suitable for scenarios where speed characteristics are not obvious (such as low-speed drones).

[0022] Furthermore, if the moving object is identified as an unauthorized drone flying illegally, then information recognition is performed on the moving image, specifically as follows:

[0023] If the moving object is identified as an unauthorized drone, the location information of the moving object is determined based on the echo information. When the location information is within a preset area range in the monitoring area, the moving image is used for information recognition to focus on and track the unauthorized drone.

[0024] By limiting the spatial range of fine processing (such as the clear recognition area), focusing and tracking are not initiated for distant targets that do not require tracing, thus reducing the system's computing power consumption; avoiding image blurring caused by focusing at a distance, and ensuring that focusing and recognition are only performed within the effective imaging range, thereby improving resource utilization efficiency.

[0025] Furthermore, the drone information identification method further includes:

[0026] The position prediction of the moving object's image data is performed using a Kalman filter algorithm to obtain the position prediction result. The imaging angles of the first imaging sensor and the second imaging sensor are then controlled based on the position prediction structure.

[0027] This method predicts the future position of the drone based on historical motion data, adjusts the gimbal angle of the imaging sensor in advance, and compensates for the response delay of the visual sensor. When the drone moves quickly or changes direction suddenly, it reduces the loss of field of view caused by "lagging adjustment" and enhances the stability and continuity of dual-sensor tracking.

[0028] Another embodiment of the present invention provides an unmanned aerial vehicle (UAV) information identification system, including: an UAV identification module, a computing module, a focusing and tracking module, and an information identification module;

[0029] The drone identification module is used to acquire motion images of moving objects within the monitoring area and identify the motion features of the moving objects in the motion images. If the moving object is identified as an unauthorized drone, the module performs information recognition on the motion image to obtain the drone type and attitude information of the unauthorized drone. Based on the drone type and attitude information, the module determines the direction information of the device identification mark of the unauthorized drone on the unauthorized drone.

[0030] The calculation module is used to calculate a first distance and a first pose relationship between the first imaging sensor and the unmanned aerial vehicle (UAV) based on the motion image, and to calculate a second distance and a second pose relationship between the second imaging sensor and the UAV. It also calculates a first angle between the first imaging sensor and the location of the device identification mark on the UAV based on the first pose relationship and the direction information, and calculates a second angle between the second imaging sensor and the location of the device identification mark on the UAV based on the second pose relationship and the direction information.

[0031] The focusing and tracking module is used to focus on the device identification of the unmanned aerial vehicle (UAV) through the first imaging sensor when the first distance is less than or equal to the second distance, or the first included angle is less than or equal to the second included angle, and to track the unmanned aerial vehicle (UAV) through the second imaging sensor to assist the first imaging sensor in focusing.

[0032] The information recognition module is used to identify the device identity identifier after the first imaging sensor has focused on the device identity identifier to obtain the identity identifier information of the unauthorized flying drone.

[0033] This invention achieves preliminary differentiation between unauthorized drones and other moving objects (such as birds and balloons) through motion feature recognition, avoiding invalid processing. Based on the approximate direction (e.g., bottom, side) of the drone's identification device, it provides "position navigation" for subsequent precise focusing, reducing blind searching. Combining attitude information (e.g., pitch angle, roll angle) corrects the identification direction judgment, ensuring more accurate identification orientation analysis and providing a basis for focusing sensor selection. By quantifying distance (first / second distance) and angle (first / second included angle), an objective standard for focusing sensor selection is established. Sensors with closer distances (reducing focusing magnification) or smaller included angles (more complete identification field of view) focus, reducing the risk of field-of-view loss due to magnification. A second sensor tracks the overall movement, providing real-time feedback on position changes, assisting the focusing sensor in dynamically adjusting the viewing angle, resolving the contradiction of "focusing and tracking being mutually exclusive." Precise focusing achieves clear imaging of tiny identifiers (QR codes, numerical strings), ensuring complete extraction of identification information, providing a unique credential for tracing unauthorized drones, and realizing a closed loop of "detection-identification-tracing."

[0034] Furthermore, the UAV identification module includes: a monitoring module, used to monitor whether the moving object exists in the monitoring area by radar, and to obtain the echo information of the radar, to perform signal processing on the echo information to obtain a signal processing result, to determine whether the moving object exists based on the signal processing result, and if it exists, to obtain the moving image through a first imaging sensor or a second imaging sensor.

[0035] In this way, radar 1 has the advantages of long-range and wide-area monitoring, and can detect moving objects before visual sensors, reducing the invalid scanning of visual sensors; it can make a preliminary judgment on the existence of moving objects through echo information, avoiding the waste of visual sensors in areas without targets, and improving the system response efficiency.

[0036] Another embodiment of the present invention provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of the UAV information identification method of the present invention.

[0037] Another embodiment of the present invention also provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform the steps of the UAV information identification method of the present invention. Attached Figure Description

[0038] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0039] Figure 1 A schematic diagram of the structure of an unmanned aerial vehicle (UAV) information identification system provided in this embodiment of the invention;

[0040] Figure 2 This is a schematic diagram of the structure of a second imaging sensor provided in an embodiment of the present invention;

[0041] Figure 3 This is a flowchart illustrating a method for identifying unmanned aerial vehicle (UAV) information provided in an embodiment of the present invention;

[0042] Figure 4 This is a schematic diagram of another unmanned aerial vehicle (UAV) information identification system provided in an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0045] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0046] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0047] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0048] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0049] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0050] For ease of description, we first propose a method such as Figure 1 The diagram shown illustrates the structure of the unmanned aerial vehicle (UAV) information identification system. Figure 2 A schematic diagram of the structure of a second imaging sensor, the system including radar 1, first imaging sensor 2, second imaging sensor 3, gimbal 4, controller 5, mobile chassis 6 and locator 7;

[0051] The first imaging sensor 2 and the second imaging sensor 3 are respectively equipped with a gimbal 4 for driving the first imaging sensor 2 and the second imaging sensor 3 to rotate.

[0052] Alternatively, the first imaging sensor 2 is located at the radar 1, and multiple second imaging sensors 3 are provided. Each second imaging sensor 3 is equipped with a mobile chassis 6, and the gimbal 4 is fixedly connected to the mobile chassis 6. A locator 7 is installed on the mobile chassis 6. The second imaging sensors 3 are deployed circumferentially around the radar 1 at 50%-100% of the maximum imaging distance of the first imaging sensor 2.

[0053] In this system architecture, the second imaging sensor 3 can be deployed independently. Under normal circumstances, the image recognition range of the camera is smaller than the scanning range of radar 1, resulting in a large number of areas where identification relies solely on radar 1. The second imaging sensor 3 can be deployed to other areas via the mobile chassis 6, working in conjunction with radar 1 to increase the area capable of image recognition and improve the monitoring range. Since focusing on the device identification requires the cooperation of two imaging sensors, deploying it at 50%-100% of the maximum imaging distance of the first imaging sensor 2 allows for a larger image overlap between the two sensors, effectively enabling the focusing of the device identification.

[0054] The imaging range of the visual sensor is limited. The gimbal 4 can drive the first imaging sensor 2 and the second imaging sensor 3 to rotate, so that the imaging range follows the movement of the black-flying drone, ensuring that the black-flying drone is always within the imaging field of view. In addition, through Kalman filtering prediction, the black-flying drone can be better locked in the imaging field of view, so as to prevent the black-flying drone from escaping the capture of the visual sensor when it moves quickly or suddenly changes its direction of movement.

[0055] See Figure 3 To address the problem of drone target loss leading to identification failure in existing technologies, an embodiment of the present invention provides a drone information identification method, including steps S301-S304, specifically including:

[0056] Step S301: Obtain motion images of moving objects within the monitoring area, and identify the motion characteristics of the moving objects in the motion images. If the moving object is identified as an unauthorized drone, perform information recognition on the motion images to obtain the drone type and attitude information of the unauthorized drone, and determine the direction information of the device identification mark of the unauthorized drone on the unauthorized drone based on the drone type and attitude information.

[0057] In this embodiment, within the monitoring area, moving objects may include unauthorized drones, birds, floating objects (such as balloons), small aircraft, and other possible aerial moving targets. When acquiring motion images, the monitoring area is captured in real-time by the first imaging sensor 2 and the second imaging sensor 3, collecting continuous image frames to form a motion image. Both the first imaging sensor 2 and the second imaging sensor 3 can be cameras.

[0058] When identifying the motion characteristics of moving objects in moving images, target detection algorithms can be used to analyze the trajectory characteristics of the moving objects. For example, it can identify whether there are regular motion trajectories unique to UAVs, such as hovering, vertical take-off and landing, fixed-point turning, and constant-speed cruising, which can be distinguished from the irregular swinging of birds or the drifting trajectory of floating objects. Motion speed characteristics can be analyzed by calculating the positional changes of moving objects in adjacent frames to obtain their movement speed. By comparing this speed with the preset speed range of unauthorized UAVs (such as 5-20 m / s), targets that are not unauthorized UAVs with excessively low speeds (such as slowly floating balloons) or excessively high speeds (such as small aircraft flying by at high speeds) can be excluded. Appearance motion characteristics can also be combined, such as identifying whether the moving object has dynamic appearance characteristics unique to UAVs, such as rotor rotation and stable fuselage attitude, to further distinguish it from other moving objects.

[0059] If the moving object is determined to be an unauthorized drone through feature recognition, the appearance features of the drone are extracted through image analysis, including the fuselage outline (such as folding type, fixed type), number of rotors (4 rotors, 6 rotors), brand logo and color scheme, etc. The appearance features are compared with the drone brand-model feature library pre-stored in the controller 5 to obtain the type information of the unauthorized drone. Then, the tilt state of the drone in the image is analyzed through attitude estimation algorithm to obtain attitude information such as pitch angle (forward and backward tilt angle of the fuselage), roll angle (left and right tilt angle of the fuselage), and yaw angle (horizontal rotation angle of the fuselage).

[0060] Based on the identified type information of the unauthorized drone, the database stored in the controller 5 contains the drone brand, model and device identification label pasting location. The preset pasting location of the device identification label for this model of drone is determined (e.g., most models are pasted at the bottom center of the fuselage, while some models are pasted on the side or top of the arm). The orientation of the label is corrected by combining its attitude information, and finally the directional information of the device identification label on the unauthorized drone (e.g., bottom orientation, side orientation, etc.) is determined.

[0061] Step S302: Calculate the first distance and first pose relationship between the first imaging sensor and the unmanned aerial vehicle (UAV) based on the motion image, and calculate the second distance and second pose relationship between the second imaging sensor and the UAV. Calculate the first angle between the first imaging sensor and the location of the device identification mark on the UAV based on the first pose relationship and the direction information, and calculate the second angle between the second imaging sensor and the location of the device identification mark on the UAV based on the second pose relationship and the direction information.

[0062] In this embodiment, frame analysis is performed on the moving images to extract the pixel coordinates and outline dimensions of the unauthorized drone in the images captured by the first imaging sensor 2 and the second imaging sensor 3. Combining the preset parameters (such as focal length and imaging resolution) of the first imaging sensor 2 and the second imaging sensor 3, an image ranging algorithm is used to calculate the straight-line distance (i.e., the first distance) between the first imaging sensor 2 and the unauthorized drone, and the straight-line distance (i.e., the second distance) between the second imaging sensor 3 and the unauthorized drone.

[0063] Simultaneously, based on the real-time coordinate position of the first imaging sensor 2 and the real-time coordinate position fed back by the locator 7 in the moving chassis 6 of the second imaging sensor 3, combined with the real-time coordinate position of the black-flying UAV in the two motion images, the first pose relationship between the first imaging sensor 2 and the black-flying UAV (including relative three-dimensional coordinates and the viewing angle direction of the first imaging sensor 2) and the second pose relationship between the second imaging sensor 3 and the black-flying UAV (including relative three-dimensional coordinates and the viewing angle direction of the second imaging sensor 3) are determined. The viewing angle direction is calibrated with the assistance of rotation angle data driven by the gimbal 4.

[0064] Based on the directional information of the identified unauthorized drone equipment, the normal direction of the surface containing the equipment identification is obtained (e.g., the normal direction of the bottom identification is vertically downward, and the normal direction of the side identification is horizontally outward). Combining this with the viewing direction of the first imaging sensor 2 in the first pose relationship, the angle between this viewing direction and the normal direction of the surface containing the equipment identification is calculated, and this angle is taken as the first angle. Similarly, combining this with the viewing direction of the second imaging sensor 3 in the second pose relationship, the angle between this viewing direction and the normal direction of the surface containing the equipment identification is calculated, and this angle is taken as the second angle.

[0065] Step S303: When the first distance is less than or equal to the second distance, or the first included angle is less than or equal to the second included angle, the first imaging sensor focuses on the device identification of the unauthorized drone, and the second imaging sensor tracks the unauthorized drone to assist the first imaging sensor in focusing.

[0066] In this embodiment, the controller 5 compares the calculated first distance with the second distance and the first angle with the second angle in real time. When the first distance between the first imaging sensor 2 and the unmanned aerial vehicle (UAV) is less than or equal to the second distance between the second imaging sensor 3 and the UAV, it indicates that the first imaging sensor 2 is closer to the target. In this case, the device identification can be clearly captured without excessive optical magnification, reducing the problem of reduced field of view caused by magnification. Alternatively, when the first angle between the first imaging sensor 2 and the device identification is less than or equal to the second angle between the second imaging sensor 3 and the device identification, it indicates that the viewing angle of the first imaging sensor 2 is closer to the orientation of the identification, reducing the probability of the identification being obscured by the drone body and making it easier to completely cover the identification area. After any of the above conditions are met, the controller 5 sends a focusing command to the first imaging sensor 2. The first imaging sensor 2 adjusts the shooting angle through its onboard gimbal 4, pointing the lens at the area where the UAV device identification is located (e.g., locking the bottom or side based on the previously determined direction information), and activates the optical magnification function (dynamically adjusting the magnification according to the distance, usually 2-3 times) until the device identification is clearly presented in the image, completing the focusing. Simultaneously, controller 5 controls the second imaging sensor 3 to switch to tracking mode. The second imaging sensor 3 adjusts the imaging angle in real time via gimbal 4 to keep the entire unmanned drone in the center of its field of view, continuously acquiring images of the drone's overall movement, extracting dynamic information such as real-time position changes and attitude adjustments (e.g., pitch and roll angle changes), and feeding this information back to controller 5 in real time. Based on the tracking data from the second imaging sensor 3, controller 5 can use a Kalman filter algorithm to predict the drone's movement trend and send angle adjustment commands to the first imaging sensor 2 in advance. This helps the first imaging sensor 2 dynamically correct its focusing angle, preventing the device identification from falling out of the first imaging sensor 2's field of view due to the drone's rapid movement or sudden change of direction, thus ensuring the stability and continuity of the focusing process.

[0067] Furthermore, since drones maintain a certain tilt during flight, there may be situations where the identification of unauthorized drones is difficult to detect from certain angles. When the second imaging sensor 3 can be deployed to other areas via the mobile chassis 6, the separation of the second imaging sensor 3 and the first imaging sensor 2 will create two perspectives in different directions. Therefore, when only image tracking is performed, the brand and model of the unauthorized drone can be determined based on the detection information, thus revealing the location of the identification. By judging the drone's flight attitude, the imaging sensor facing which the identification is located can be determined, and focusing on the identification based on that sensor reduces the probability of the identification being obscured by other parts of the drone. In addition, when the unauthorized drone is located between the second imaging sensor 3 and the first imaging sensor 2, allocating image tracking and identification focusing tasks based on the distance between the drone and the imaging sensors can improve the image clarity during the focusing task. Therefore, when the unauthorized drone is positioned between the second imaging sensor 3 and the first imaging sensor 2, if the drone's identification tag is facing the first imaging sensor 2, the first imaging sensor 2 focuses on the identification tag, and the second imaging sensor 3 maintains image tracking; if the drone's identification tag is facing the second imaging sensor 3, the second imaging sensor 3 focuses on the identification tag, and the first imaging sensor 2 maintains image tracking. When the drone is hovering in a tilt-free state, if the drone is closer to the first imaging sensor 2, the first imaging sensor 2 focuses on the identification tag, and the second imaging sensor 3 maintains image tracking; if the drone is closer to the second imaging sensor 3, the second imaging sensor 3 focuses on the identification tag, and the first imaging sensor 2 maintains image tracking.

[0068] Step S304: After the first imaging sensor has focused on the device identification mark, it identifies the device identification mark to obtain the identification mark information of the unauthorized drone.

[0069] In this embodiment, once the first imaging sensor 2 has focused on the device identification mark through gimbal adjustment and optical magnification by the gimbal 4, and the device identification mark (such as a QR code, barcode, or string of numbers) is clearly identifiable in the image, the controller 5 initiates the identification recognition process. The first imaging sensor 2 transmits the focused high-definition image frame to the controller 5, which then calls a preset image recognition algorithm (such as a QR code decoding algorithm, a barcode recognition module, or optical character recognition (OCR) technology) to analyze the device identification mark in the image. During the analysis, the image recognition algorithm can first preprocess the image, including noise reduction, contrast enhancement, and geometric distortion correction, to ensure complete extraction of the identification features; then, it performs specific identification based on the identification type—if it is a QR code or barcode, the decoding algorithm reads the encoded information contained therein; if it is a string of numbers or other character identifiers, it is converted into text information using OCR technology. The identified encoded or text information is the unique identification information of the unauthorized drone. The controller 5 associates and matches this information with a pre-stored drone identification database, ultimately completing the traceability and confirmation of the unauthorized drone's identity. During this process, the second imaging sensor 3 continuously tracks the unmanned aerial vehicle (UAV) and provides real-time feedback on its motion status, ensuring that the first imaging sensor 2 maintains a stable focus on the marker during identification, thus avoiding identification interruption due to minor movements of the UAV.

[0070] As an example of an embodiment of the present invention, the step of acquiring the motion image of a moving object within the monitoring area specifically includes:

[0071] The system monitors the presence of the moving object within the monitoring area using radar and acquires the radar echo information. The echo information is then processed to obtain a signal processing result. Based on the signal processing result, it is determined whether a moving object exists. If so, the moving image is acquired using a first imaging sensor or a second imaging sensor.

[0072] In this embodiment, radar 1 continuously scans the monitoring area, emitting electromagnetic waves and receiving the reflected echo information. Controller 5 processes the echo information to extract information related to moving targets, forming a signal processing result. Controller 5 determines whether a moving object exists within the monitoring area based on the signal processing result. If so, controller 5 selects either the first imaging sensor 2 or the second imaging sensor 3 to start operation based on the moving object's position information. Controller 5 controls the selected imaging sensor to adjust its angle via its mounted gimbal 4, aiming at the moving object and initiating image capture, acquiring continuous image frames of the moving object to obtain a motion image of the object.

[0073] As an example of an embodiment of the present invention, the step of acquiring a motion image of a moving object within a monitoring area, wherein the motion features include speed features and motion trajectory features, specifically:

[0074] The echo information is subjected to noise filtering and clutter suppression to obtain target echo information. The target echo information is then clustered to identify various moving objects, and the echo intensity of each moving object is determined based on the target echo information. Combining the echo intensity, a convolutional neural network is used to identify the motion characteristics of each moving object. If the identified motion trajectory characteristics of the target moving object are those of a UAV flight trajectory, or if the speed value of the target moving object is greater than a preset threshold, then the target moving object is identified as the unauthorized UAV, and a motion image of the target moving object is acquired.

[0075] Among them, the convolutional neural network is used for training based on the flight trajectory samples of unauthorized drones. The learning rate of the convolutional neural network can be 0.01-0.05, and the number of training samples is greater than 1800. The convolutional neural network is used to determine whether the moving target is an unauthorized drone based on the motion trajectory of the input moving object.

[0076] In this embodiment, the controller 5 performs signal processing on the acquired echo information. First, it filters out invalid signals such as environmental electromagnetic interference and ground reflection through noise filtering and clutter suppression algorithms to extract valid target echo information. Then, it performs grouping and clustering processing based on the echo information to identify different moving objects that may exist in the monitoring area, eliminate overlap or interference between targets, and calculate the distance and azimuth of the moving objects by measuring the propagation time of the radar 1 signal to form the signal processing result.

[0077] The controller 5 determines whether a moving object exists within the monitoring area based on the signal processing results: if no valid target echo is identified after clustering or the target characteristics do not match the characteristics of a moving object, it is determined that there is no moving object; if a target echo with continuous motion characteristics is identified (such as the presence of Doppler frequency shift), it is determined that a moving object exists. Once a moving object is determined to exist, the controller 5 sends a start command to the first imaging sensor 2 or the second imaging sensor 3 based on the azimuth and distance information of the moving object. If the moving object is closer to the first imaging sensor 2 or is within its main monitoring range, the controller 5 controls the first imaging sensor 2 to adjust its angle to align with the moving object via the gimbal 4, starts shooting, and acquires continuous image frames to form a motion image; if the moving object is closer to the second imaging sensor 3 or is within its deployed monitoring area, the controller 5 controls the second imaging sensor 3 to adjust its position with the assistance of the mobile chassis 6 and the locator 7, and, in conjunction with the gimbal 4, aligns with the target, acquires motion images, and transmits them to the controller 5 in real time.

[0078] As an example of an embodiment of the present invention, if the moving object is identified as an unauthorized drone flying in the wild, then information recognition is performed on the moving image, specifically as follows:

[0079] If the moving object is identified as an unauthorized drone, the location information of the moving object is determined based on the echo information. When the location information is within a preset area range in the monitoring area, the moving image is used for information recognition to focus on and track the unauthorized drone.

[0080] In this embodiment, if the moving object is determined to be an unauthorized drone through image recognition and motion feature analysis, the controller 5 calculates the real-time position coordinates of the unauthorized drone (including the straight-line distance and azimuth angle from radar 1) based on the echo information of radar 1, and determines its specific position information within the monitoring area. The controller 5 has a preset area range value (such as an airspace range at a certain distance from the core monitoring area) for defining the fine identification range, and compares the position information of the unauthorized drone with the preset area range.

[0081] When the location coordinates of the unauthorized drone fall within the preset area, it indicates that it has entered the distance range where device identification can be effectively performed. The controller 5 then triggers the information recognition process for the moving image: on the one hand, the video stream of the moving image is processed by the target detection algorithm to extract the appearance features of the unauthorized drone (such as brand logo, fuselage outline, number of rotors) to determine its model. At the same time, the tilt angle of the drone in the image is analyzed to obtain attitude information such as pitch angle and roll angle. On the other hand, based on the identified model information, the pre-stored database is queried to determine the pasting position and orientation of the device identification mark. This provides accurate target positioning basis for the first imaging sensor 2 to focus on the device identification mark and the second imaging sensor 3 to track the overall movement of the drone, ensuring the coordinated and efficient focusing and tracking process.

[0082] As an example of an embodiment of the present invention, the UAV information identification method further includes:

[0083] The position prediction of the moving object's image data is performed using a Kalman filter algorithm to obtain the position prediction result. The imaging angles of the first imaging sensor and the second imaging sensor are then controlled based on the position prediction structure.

[0084] In this embodiment, Kalman filtering is a recursive algorithm used to dynamically estimate the system state from noisy observation data. In visual target tracking, it combines motion models and observation data to predict and correct the target's position, velocity, and other states. Kalman filtering involves two iterative phases: prediction and update. Prediction is based on the target's motion model, including constant velocity and constant acceleration predictions of the next moment's state and uncertainties. New observation data, such as target detection boxes, is used to correct the predicted values ​​and reduce uncertainty. The gimbal is adjusted according to the motion model's state and uncertainties to keep the target centered in the visual analysis module's imaging.

[0085] Specifically, controller 5 continuously receives moving object image data acquired by the first imaging sensor 2 and the second imaging sensor 3, extracting dynamic features of the moving object in each frame, such as pixel coordinates, speed, and direction. This real-time image data is then input into a pre-configured Kalman filter algorithm model. This model establishes a motion model (such as a uniform or uniformly accelerated motion model) based on the moving object's historical trajectory. Through a "prediction-update" iterative process, the theoretical position coordinates of the moving object at the next moment are calculated, yielding a position prediction result, including the predicted pixel coordinates and the possible range of position deviation.

[0086] Based on the position prediction results output by the Kalman filter, controller 5 sends angle adjustment commands to the gimbals 4 of the first imaging sensor 2 and the second imaging sensor 3. For the first imaging sensor 2, if it is in the state of focusing on the device identification mark, controller 5 fine-tunes the horizontal and vertical rotation angles of the gimbal 4 based on the predicted position to ensure that the device identification mark is always in the center of the imaging field of view. For the second imaging sensor 3, controller 5 drives the gimbal 4 to adjust its direction in real time based on the predicted position to maintain stable tracking of the moving object in the shooting frame. When the moving object moves rapidly or suddenly changes direction, the Kalman filter algorithm dynamically corrects the prediction model and outputs the position change trend in advance, enabling the gimbal 4 to adjust the imaging angle in advance, preventing the moving object from leaving the imaging field of view and ensuring the continuity of the focusing and tracking process.

[0087] This embodiment uses Kalman filtering prediction to better lock the unauthorized drone within the imaging field of view, so as to prevent the unauthorized drone from escaping the capture of the visual sensor when it moves quickly or suddenly changes direction.

[0088] like Figure 4 As shown, based on the above method embodiment, another corresponding system embodiment is provided; one embodiment of the present invention provides a UAV information identification system 400, including: a UAV identification module 401, a calculation module 402, a focusing and tracking module 403, and an information identification module 404;

[0089] The drone identification module 401 is used to acquire motion images of moving objects within the monitoring area and identify the motion features of the moving objects in the motion images. If the moving object is identified as an unauthorized drone, the module performs information recognition on the motion image to obtain the drone type and attitude information of the unauthorized drone, and determines the direction information of the device identification mark of the unauthorized drone on the unauthorized drone based on the drone type and attitude information.

[0090] The calculation module 402 is used to calculate the first distance and first pose relationship between the first imaging sensor and the unmanned aerial vehicle (UAV) based on the motion image, and to calculate the second distance and second pose relationship between the second imaging sensor and the UAV. Based on the first pose relationship and the direction information, it calculates the first angle between the first imaging sensor and the location of the device identification mark on the UAV, and calculates the second angle between the second imaging sensor and the location of the device identification mark on the UAV based on the second pose relationship and the direction information.

[0091] The focusing and tracking module 403 is used to focus on the device identification of the unmanned aerial vehicle (UAV) through the first imaging sensor when the first distance is less than or equal to the second distance, or the first included angle is less than or equal to the second included angle, and to track the unmanned aerial vehicle (UAV) through the second imaging sensor to assist the first imaging sensor in focusing.

[0092] The information recognition module 404 is used to recognize the device identity identifier after the first imaging sensor has focused on the device identity identifier to obtain the identity identifier information of the unauthorized flying drone.

[0093] As an example of an embodiment of the present invention, the UAV identification module 401 includes: a monitoring module, used to monitor whether the moving object exists in the monitoring area by radar, and to obtain the echo information of the radar, to perform signal processing on the echo information to obtain a signal processing result, to determine whether the moving object exists based on the signal processing result, and if it exists, to obtain the moving image by a first imaging sensor or a second imaging sensor.

[0094] It is understood that the above system item embodiments correspond to the method item embodiments of the present invention, and can implement the UAV information identification method X provided by any of the above method item embodiments of the present invention.

[0095] It should be noted that the system embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the system embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0096] For ease of description and brevity, the system embodiments of the present invention include all the implementation methods described in the above embodiments of the UAV-based information recognition method, and will not be repeated here.

[0097] Based on the above embodiments of the UAV information identification method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the UAV information identification method of any embodiment of the present invention.

[0098] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0099] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0100] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0101] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the UAV information identification method described in any of the above-described method embodiments of the present invention.

[0102] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0103] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for identifying information about unmanned aerial vehicles (UAVs), characterized in that, include: The motion image of the moving object in the monitoring area is acquired, and the motion characteristics of the moving object in the motion image are identified. If the moving object is identified as an unauthorized flying drone, the motion image is identified to obtain the drone type and attitude information of the unauthorized flying drone. Based on the drone type and attitude information, the direction information of the device identification mark of the unauthorized flying drone on the unauthorized flying drone is determined. Calculate the first distance and first pose relationship between the first imaging sensor and the unmanned aerial vehicle (UAV) based on the motion image, and calculate the second distance and second pose relationship between the second imaging sensor and the UAV. Calculate the first angle between the first imaging sensor and the location of the device identification mark on the UAV based on the first pose relationship and the direction information, and calculate the second angle between the second imaging sensor and the location of the device identification mark on the UAV based on the second pose relationship and the direction information. When the first distance is less than or equal to the second distance, or the first included angle is less than or equal to the second included angle, the first imaging sensor focuses on the device identification of the unauthorized drone, and the second imaging sensor tracks the unauthorized drone to assist the first imaging sensor in focusing; Once the first imaging sensor has focused on the device identification identifier, it identifies the device identification identifier to obtain the identification information of the unauthorized drone.

2. The UAV information identification method as described in claim 1, characterized in that, The acquisition of motion images of moving objects within the monitoring area specifically involves: The system monitors the presence of the moving object within the monitoring area using radar and acquires the radar echo information. The echo information is then processed to obtain a signal processing result. Based on the signal processing result, it is determined whether a moving object exists. If so, the moving image is acquired using a first imaging sensor or a second imaging sensor.

3. The UAV information identification method as described in claim 2, characterized in that, The acquisition of motion images of moving objects within the monitoring area, wherein the motion features include velocity features and motion trajectory features, specifically: The echo information is subjected to noise filtering and clutter suppression to obtain target echo information. The target echo information is then clustered to identify various moving objects. The echo intensity of each moving object is determined based on the target echo information. Based on the echo intensity, the motion characteristics of each moving object are identified by a convolutional neural network. If the motion trajectory characteristics of the target moving object are identified as those of a drone flight trajectory, or if the speed value of the target moving object is greater than a preset threshold, then the target moving object is determined to be the unauthorized drone, and the motion image of the target moving object is acquired.

4. The UAV information identification method as described in claim 2, characterized in that, The acquisition of motion images of moving objects within the monitoring area specifically involves: The echo information is subjected to noise filtering and clutter suppression to obtain target echo information. The target echo information is then clustered. If multiple moving objects are identified, the motion trajectory of each moving object is identified using a convolutional neural network. If the motion trajectory feature of the target moving object is identified as the flight trajectory feature of a UAV, the target moving object is determined to be the unauthorized UAV, and the motion image of the target moving object is acquired.

5. The UAV information identification method as described in claim 3, characterized in that, If the moving object is identified as an unauthorized drone flying illegally, then information recognition is performed on the moving image, specifically as follows: If the moving object is identified as an unauthorized drone, the location information of the moving object is determined based on the echo information. When the location information is within a preset area range in the monitoring area, the moving image is used for information recognition to focus on and track the unauthorized drone.

6. The UAV information identification method as described in claims 1-5, characterized in that, The UAV information identification method further includes: The position prediction of the moving object's image data is performed using a Kalman filter algorithm to obtain the position prediction result. The imaging angles of the first imaging sensor and the second imaging sensor are then controlled based on the position prediction structure.

7. A drone information identification system, characterized in that, include: The drone identification module, computing module, focusing and tracking module, and information recognition module; The drone identification module is used to acquire motion images of moving objects within the monitoring area and identify the motion features of the moving objects in the motion images. If the moving object is identified as an unauthorized drone, the module performs information recognition on the motion image to obtain the drone type and attitude information of the unauthorized drone. Based on the drone type and attitude information, the module determines the direction information of the device identification mark of the unauthorized drone on the unauthorized drone. The calculation module is used to calculate a first distance and a first pose relationship between the first imaging sensor and the unmanned aerial vehicle (UAV) based on the motion image, and to calculate a second distance and a second pose relationship between the second imaging sensor and the UAV. It also calculates a first angle between the first imaging sensor and the location of the device identification mark on the UAV based on the first pose relationship and the direction information, and calculates a second angle between the second imaging sensor and the location of the device identification mark on the UAV based on the second pose relationship and the direction information. The focusing and tracking module is used to focus on the device identification of the unmanned aerial vehicle (UAV) through the first imaging sensor when the first distance is less than or equal to the second distance, or the first included angle is less than or equal to the second included angle, and to track the unmanned aerial vehicle (UAV) through the second imaging sensor to assist the first imaging sensor in focusing. The information recognition module is used to identify the device identity identifier after the first imaging sensor has focused on the device identity identifier to obtain the identity identifier information of the unauthorized flying drone.

8. A drone information identification system, characterized in that, The UAV identification module includes a monitoring module, which is used to monitor whether the moving object exists in the monitoring area by radar, and to obtain the echo information of the radar, to process the echo information to obtain the signal processing result, and to determine whether the moving object exists based on the signal processing result. If it exists, the moving image is obtained by a first imaging sensor or a second imaging sensor.

9. A terminal device, characterized in that, The device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the UAV information identification method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the unmanned aerial vehicle information identification method as described in any one of claims 1-6.

Images