Speed measurement method and inspection system based on dynamic zooming of unmanned aerial vehicle

Abstract

The invention provides a speed measurement method based on dynamic zooming of an unmanned aerial vehicle and an inspection system. The method comprises the following steps: acquiring a video stream acquired by the unmanned aerial vehicle in real time through a camera for dynamically adjusting focal length parameters; inputting the video stream into the detection model frame by frame; outputting a target detection frame set of the corresponding current frame; performing cross-frame association on the detection frames of the adjacent frames through a ByteTrack algorithm, and distributing a unique instance identifier for the same target; acquiring real-time parameters of the unmanned aerial vehicle, establishing a real pavement coordinate system and a pixel imaging coordinate system, and performing coordinate system projection mapping; and calculating the real speed of the target. The method has the beneficial effects that the coordinate mapping error is greatly reduced through dynamic parameter fusion and unmanned aerial vehicle displacement compensation, and particularly, the improvement is obvious in a pitch angle or zoom scene; through introduction of central point position transformation, image center offset caused by digital zooming is dynamically calibrated, and pixel-physical coordinate mapping distortion caused by pure image cutting or interpolation amplification is effectively eliminated.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic management, and in particular relates to a speed measurement method and inspection system based on dynamic zoom of an unmanned aerial vehicle. Background Art

[0002] Existing drone-based vehicle speed measurement technology analyzes the image coordinates of targets in a video stream and maps them to real-world road locations based on the drone's flight parameters to calculate the target's speed. However, existing technologies have significant shortcomings: traditional detection models (such as two-stage detectors) suffer from severe computational redundancy and are unable to meet real-time requirements; speed estimation models fail to fully account for projection errors caused by drone dynamic parameters (such as pitch angle changes and zoom operations); and existing geometric coordinate mapping methods rely on the assumption that the drone's flight plane is strictly parallel to the road, making them susceptible to angular deviations and the drone's own displacement in practical applications. Furthermore, existing technologies fail to effectively distinguish between the physical effects of optical and digital zoom, significantly reducing coordinate mapping stability in zoom scenarios. Therefore, a real-time speed measurement method and system that integrates drone dynamic parameters, an adaptive zoom mechanism, and compensates for the drone's own motion is urgently needed to address these issues. Summary of the Invention

[0003] In order to solve the above technical problems, the present invention provides a speed measurement method and inspection system based on dynamic zoom of a drone, which is particularly suitable for speed measurement and monitoring of targets on the road through a drone.

[0004] The technical solution adopted by the present invention is: in the first aspect, a method for measuring speed based on dynamic zoom of a drone is provided, comprising:

[0005] Obtain the real-time video stream collected by the drone's camera with dynamically adjusted focal length parameters;

[0006] Input the video stream frame by frame into the detection model;

[0007] Output the corresponding target detection frame set of the current frame;

[0008] The ByteTrack algorithm is used to associate the detection boxes of adjacent frames across frames and assign a unique instance identifier to the same target.

[0009] Obtain the real-time parameters of the drone, establish the real road coordinate system and pixel imaging coordinate system, and perform coordinate system projection mapping;

[0010] Calculate the target's true velocity.

[0011] Furthermore, the detection includes a YOLOv8s model.

[0012] Furthermore, the real-time parameters of the UAV include the UAV altitude, the UAV pitch angle, the camera viewing angle, the image pixel width, the video frame rate, and the sensor maximum image aspect ratio.

[0013] Furthermore, establishing the real road surface coordinate system and the pixel imaging coordinate system and performing coordinate system projection mapping includes the following steps:

[0014] The real-time projection point of the drone on the ground is taken as the origin, the forward direction of the drone is taken as the y-axis, and the The ground normal vector of the point is the z axis, the plane The normal vector of is the x-axis, and the real road coordinate system is established;

[0015] The normal vector of the plane where the sensor at the center of the drone lens is located is Axis, the intersection of this axis and the plane where the sensor is located is the origin , the direction of sensor width is Axis, the direction where the sensor is high is Axis establishes pixel imaging coordinate system;

[0016] Through the equation , Perform coordinate system projection mapping, where ( ) is the coordinate of the target in the pixel imaging coordinate system, is the coordinate of the target in the real road coordinate system, is the maximum image aspect ratio of the sensor, is the pitch angle of the drone, is the image pixel width, is the camera perspective, The flight altitude of the drone;

[0017] Through the equation Get the target speed at time t ,in is the position of the UAV at time t in the real road coordinate system, is the position of the UAV at time t-1 in the real road coordinate system, and the flight speed scalar of the UAV at time t is , the video frame rate is f.

[0018] Furthermore, through equation Suppress the random noise generated by the jitter during the flight of the drone and obtain the suppressed speed of the target at time t ,in is the relative coordinate vector of the target to the UAV at time t obtained by the coordinate transformation algorithm, is the relative coordinate vector of the target to the UAV at time tk obtained by the coordinate transformation algorithm, and the flight speed vector of the UAV at time ti is , k is the time interval, k is 10.

[0019] Furthermore, the drone dynamically adjusts the focal length parameters in two ways: optical zoom and digital zoom. When digital zoom is used, the detection frame coordinates are scaled according to the zoom ratio; when optical zoom is used, the equivalent focal length parameters of the lens are updated.

[0020] Furthermore, when digital zoom is used, the equation Add center point position transformation to calibrate target velocity calculation, where is the zoom magnification, The minimum magnification of different lenses.

[0021] In a second aspect, a speed measurement inspection system based on a dynamic zoom of a drone is provided, comprising:

[0022] Monitoring module, used to monitor targets on the road through drones and obtain video streams;

[0023] The target classification and recognition module processes the video stream to identify the types of targets, including cars, buses, trucks, and pedestrians;

[0024] The road congestion detection module receives the video stream and the target type to calculate the number of cars, buses, and trucks, and then determines whether the road is congested and outputs the result;

[0025] The vehicle speed calculation module receives the video stream, road congestion conditions, and target types to calculate the vehicle speed. When the road is not congested, it marks and alerts vehicles that are below the speed limit.

[0026] The pedestrian detection module is used to receive video streams and target types for pedestrian detection, and mark and warn when pedestrians are detected on the road.

[0027] In a third aspect, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the speed measurement method based on dynamic zoom of a drone provided in the present disclosure.

[0028] In a fourth aspect, a computer program product is provided, including a computer program / instruction, which, when executed by a processor, implements the speed measurement method based on dynamic zoom of a drone provided in the present disclosure.

[0029] The advantages and positive effects of the present invention are as follows: due to the adoption of the above technical solution, a lightweight YOLOv8s detection model and a ByteTrack tracking algorithm are selected to reduce computational complexity and video memory usage, achieve high frame rate processing capability, effectively reduce ID jumps in road scenes with little occlusion and continuous target displacement, and enhance trajectory continuity; dynamic parameter fusion and drone displacement compensation greatly reduce coordinate mapping errors, especially in pitch angle or zoom scenarios; support stable speed measurement of drones under wide-area flight conditions, and significantly improve the applicability of the system; by introducing center point position transformation, dynamically calibrate the image center offset caused by digital zoom, effectively eliminate pixel-physical coordinate mapping distortion caused by simple image cropping or interpolation amplification, and avoid the risk of mispositioning; through the module cascade design within the system, flexible switching between wide-angle monitoring and detailed targets is achieved, while ensuring global traffic situation awareness, meeting local high-precision speed measurement requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 1 is a flow chart of a method for measuring speed based on dynamic zoom of a drone according to an embodiment of the present invention;

[0031] Figure 2 is a schematic diagram of a real road surface coordinate system according to an embodiment of the present invention;

[0032] Figure 3 is a schematic diagram of a pixel imaging coordinate system according to an embodiment of the present invention;

[0033] Figure 4 is a schematic diagram of the wide plane geometric structure of an embodiment of the present invention;

[0034] Figure 5 Schematic diagram of the high-plane geometric structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The present disclosure is described more fully below with reference to the accompanying drawings, which illustrate exemplary embodiments of the present disclosure. The technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present disclosure.

[0036] like Figure 1 As shown, the present invention provides a speed measurement method based on dynamic zoom of a drone, comprising:

[0037] S100, obtaining a video stream collected in real time by a camera of the drone that dynamically adjusts focal length parameters;

[0038] S200, inputting the video stream into the detection model frame by frame;

[0039] S300, outputting a corresponding target detection frame set of the current frame;

[0040] S400, performing cross-frame association on detection frames of adjacent frames using a ByteTrack algorithm, and assigning a unique instance identifier to the same target;

[0041] S500: Acquire real-time parameters of the UAV, establish a real road coordinate system and a pixel imaging coordinate system, and perform coordinate system projection mapping;

[0042] S600: Calculate the true speed of the target.

[0043] The above method uses a lightweight YOLOv8s detection model and a ByteTrack tracking algorithm to reduce computational complexity and video memory usage, achieve high frame rate processing capabilities, and effectively reduce ID jumps and enhance trajectory consistency in road scenes with little occlusion and continuous target displacement.

[0044] In one embodiment, the detection model includes a YOLOv8s model.

[0045] In order to solve the problem that the target speed cannot be directly calculated from the video stream captured by the drone, an implementation method is provided in this embodiment.

[0046] like Figure 2-Figure 3 As shown, in one embodiment, establishing a real road surface coordinate system and a pixel imaging coordinate system and performing coordinate system projection mapping includes the following steps:

[0047] The real-time projection point of the drone on the ground is taken as the origin, the forward direction of the drone is taken as the y-axis, and the The ground normal vector of the point is the z axis, the plane The normal vector of is the x-axis, and the real road coordinate system is established;

[0048] The normal vector of the plane where the sensor at the center of the drone lens is located is Axis, the intersection of this axis and the plane where the sensor is located is the origin , the direction of sensor width is Axis, the direction where the sensor is high is Axis establishes pixel imaging coordinate system;

[0049] Pass the equation , Perform coordinate system projection mapping, where ( ) is the coordinate of the target in the pixel imaging coordinate system, is the coordinate of the target in the real road coordinate system, is the maximum image aspect ratio of the sensor, is the pitch angle of the drone, is the image pixel width, is the camera perspective, The flight altitude of the drone;

[0050] Through the equation Get the speed of the target (the target observed by the drone) at time t ,in is the position of the UAV at time t in the real road coordinate system, and the scalar of the UAV's flight speed at time t is , the video frame rate is f.

[0051] Using this method, dynamic parameter fusion and drone displacement compensation significantly reduce coordinate mapping errors, especially in pitch angle or zoom scenarios. It also supports stable speed measurement under wide-area drone flight conditions (altitude changes, pitch angle fluctuations), significantly improving the system's applicability.

[0052] In one embodiment, in order to suppress the random noise generated by the jitter of the drone during flight, that is, to estimate the target speed more accurately after considering the impact of the drone jitter, the equation Get the target speed at time t (i.e., the estimated speed after suppressing the drone's vibration), where is the relative coordinate vector of the target to the UAV at time t obtained by the coordinate transformation algorithm, is the relative coordinate vector of the target to the UAV at time tk obtained by the coordinate transformation algorithm, and the flight speed vector of the UAV at time ti is , the vector of the UAV’s flight speed at time t is (The horizontal component is 0), k is the time interval taken, and k is 10.

[0053] In order to solve the problem that the drone selects different zoom modes according to different situations when shooting, thereby causing speed jumps, an implementation method is provided in this embodiment.

[0054] In one embodiment, the drone dynamically adjusts the focal length parameters by optical zoom and digital zoom. When digital zoom is used, the detection frame coordinates are scaled according to the zoom ratio; when optical zoom is used, the equivalent focal length parameters of the lens are updated.

[0055] The above method is used to distinguish the processing mechanisms of optical and digital zoom, thereby avoiding speed jumps caused by zoom operations.

[0056] In order to solve the problem of image center offset caused by digital zoom during drone photography, an implementation method is provided in this embodiment.

[0057] In one embodiment, when digital zoom is used, the equation Add center point position transformation to calibrate target velocity calculation, where is the zoom magnification, The minimum magnification of different lenses.

[0058] By adopting the above method, by introducing the center point position transformation, the image center offset caused by digital zoom is dynamically calibrated, effectively eliminating the pixel-physical coordinate mapping distortion caused by simple image cropping or interpolation amplification, and avoiding the risk of mispositioning.

[0059] The following describes the contents involved in the above embodiment in conjunction with a preferred embodiment.

[0060] like Figure 2-Figure 5 As shown, a DJI M300RTK drone equipped with an H20T camera is used to shoot 4K@30fps video streams in wide-angle mode by default. Data is transmitted back in real time via dual-frequency image transmission. The video stream is input frame by frame into the CMM-based YOLOv8s model, which outputs the corresponding target detection frame set for the current frame. The detection frames of adjacent frames are associated across frames using the ByteTrack algorithm, and a unique instance identifier is assigned to the same target. The real-time parameters of the drone are obtained, including image pixel width w, image pixel height h, and drone height. , UAV pitch angle , camera field of view (FOV) and the camera's maximum image aspect ratio , video frame rate f, the coordinate origin of the image is migrated, through the equation Convert the origin of the upper left corner of the detection frame to the center of the image, take the real-time projection point of the drone on the ground as the origin, and the forward direction of the drone as the y-axis. The ground normal vector of the point is the z axis, the plane The normal vector of the x-axis is used to establish the real road coordinate system; the normal vector of the plane where the sensor at the center of the drone lens is located is Axis, the intersection of this axis and the plane where the sensor is located is the origin , the direction of sensor width is Axis, the direction where the sensor is high is The pixel imaging coordinate system is established by the axis; the target position on the ground is point c, and the position on the image is point In the coordinate system xOy, the projection of point c on the y-axis is , in the image plane coordinate system midpoint exist The projection onto the axis is ,exist The projection onto the axis is According to the principle of convex lens optical imaging, it is obvious that in the real physical coordinate system and the pixel imaging coordinate system, the plane With plane Coplanar, flat surface With plane Coplanar. The target position relative to the drone is That is .exist Plane, that is, high plane, such as Figure 5 , through equation , The y coordinate of the target in the actual space coordinate system can be obtained ,exist Flat surface, i.e. wide surface, e.g. Figure 4 ,triangle With triangle Similar, can be obtained By equation , Perform coordinate system projection mapping, where is the coordinate of the target in the real road coordinate system. If the UAV remains stationary, then the equation Find the target's speed. If the drone vibrates during flight, then the equation Get the target speed at time t ,in is the relative coordinate vector of the target to the UAV obtained by the coordinate transformation algorithm, and the vector of the UAV’s flight speed is , k is the time interval, k is 10. The drone dynamically adjusts the focal length parameters by optical zoom and digital zoom. When digital zoom is used, the detection frame coordinates are scaled according to the zoom ratio; when optical zoom is used, the equivalent focal length parameters of the lens are updated. When digital zoom is used, the equation Add center point position transformation to calibrate target velocity calculation, where is the zoom magnification, The minimum magnification of different lenses.

[0061] In order to facilitate the use of the speed measurement method based on dynamic zoom of a drone provided by the present disclosure, the present disclosure also provides a speed measurement inspection system based on dynamic zoom of a drone, including:

[0062] Monitoring module, used to monitor targets on the road through drones and obtain video streams;

[0063] The target classification and recognition module processes the video stream to identify the types of targets, including cars, buses, trucks, and pedestrians;

[0064] The road congestion detection module receives the video stream and the target type to calculate the number of cars, buses, and trucks, and then determines whether the road is congested and outputs the result;

[0065] The vehicle speed calculation module receives the video stream, road congestion conditions, and target types to calculate the vehicle speed. When the road is not congested, it marks and alerts vehicles that are below the speed limit.

[0066] The pedestrian detection module is used to receive video streams and target types for pedestrian detection, and mark and warn when pedestrians are detected on the road.

[0067] With the above settings, the target classification and recognition module can accurately distinguish between cars, buses, trucks and pedestrians, which can adapt to the needs of multiple scenarios such as urban roads and highways, avoid missed detections or false detections caused by traditional methods relying on single vehicle detection, and improve the level of refined traffic management; the road congestion detection module dynamically counts the number of target vehicles, judges the congestion status in real time, provides data support for vehicle diversion and scheduling, and reduces the decline in traffic efficiency caused by congestion; the vehicle speed calculation module combines congestion status classification processing: in non-congested conditions, it focuses on monitoring ultra-low-speed vehicles and marking them in real time to avoid the risk of rear-end collisions caused by slow-moving vehicles; in congested conditions, the speed measurement frequency is reduced to save computing power and improve the efficiency of system resource utilization; the pedestrian detection module instantly marks and warns pedestrians on the road, combines the high-altitude perspective advantage of drones to avoid the blind spot problem of ground equipment, and improves safety in complex road conditions; through the module cascade design within the system, flexible switching between wide-angle monitoring and detailed targets is achieved, while ensuring global traffic situation awareness, meeting local high-precision speed measurement requirements.

[0068] The following describes the contents involved in the above embodiment in conjunction with a preferred embodiment.

[0069] The monitoring module uses a DJI M3TD drone, capturing 4K@30fps video streams at magnifications of 1-8 (the DJI M3TD drone's lens has a wide-angle zoom of 1-7x, and a telephoto zoom of 7-56x). This data is transmitted back in real time via dual-frequency image transmission. The object classification and recognition module is based on the YOLOv8s model, with the classification head extended to include, but not limited to, cars, buses, trucks, and pedestrians. The dataset is trained using BDD100K and custom drone-view augmented data. The module receives video frames from the monitoring module and outputs detection bounding boxes and categories using YOLOv8s and a deep learning model. A deep learning real-time semantic segmentation network is used to segment roads, removing non-road objects (such as trees and signs) to reduce computational load. The road congestion detection module receives video frames and object types, counting the number of target vehicles within the road segment covered by the video stream every 15 seconds. Congestion is detected when the ratio of the number of vehicles to the number of roads (or road length) exceeds a set threshold. The vehicle speed calculation module receives video frames, target type, and congestion conditions, and calculates the speed of all vehicles in the video frame. Any vehicle speeding below 70% of the minimum speed limit is marked as abnormally slow. The drone broadcasts an alarm and pushes the alert to the ground terminal for broadcast. Low-speed vehicle detection is discontinued in congested roads to save computing power. The pedestrian detection module receives video frames and target type, and detects pedestrians in vehicle-accessible areas (such as motorway lanes and non-segregated intersections) in real time. If a pedestrian's coordinates fall within the road boundary, the video stream is superimposed with a red box to mark the pedestrian. The drone also broadcasts an alarm and pushes the alert to the ground terminal for broadcast.

[0070] Based on the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0071] An electronic device includes at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the speed measurement method based on dynamic zoom of a drone provided by the present disclosure.

[0072] Electronic device is intended to refer to various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device may also refer to various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are intended to be examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0073] A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the speed measurement method based on dynamic zoom of an unmanned aerial vehicle provided by the present disclosure.

[0074] Various embodiments of the present disclosure may be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system comprising at least one programmable processor, which may be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0075] A computer program product includes a computer program / instruction. When the computer program / instruction is executed by a processor, the speed measurement method based on dynamic zoom of a drone provided by the present disclosure is implemented.

[0076] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0077] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0078] The embodiments of the present invention are described in detail above, but the contents described are only preferred embodiments of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. The speed measurement method based on the dynamic zoom of the UAV is characterized by: include: Obtain the real-time video stream collected by the drone's camera with dynamically adjusted focal length parameters; Input the video stream frame by frame into the detection model; Output the corresponding target detection frame set of the current frame; The ByteTrack algorithm is used to associate the detection boxes of adjacent frames across frames and assign a unique instance identifier to the same target. Obtain the real-time parameters of the drone, establish the real road coordinate system and pixel imaging coordinate system, and perform coordinate system projection mapping; Calculate the target's true velocity.

2. The method for measuring speed based on dynamic zoom of a drone according to claim 1, characterized in that: The detection model includes a YOLOv8s model.

3. The method for measuring speed based on dynamic zoom of a drone according to claim 1, characterized in that: The real-time parameters of the drone include drone altitude, drone pitch angle, camera viewing angle, image pixel width, video frame rate, and sensor maximum image aspect ratio.

4. The speed measurement method based on dynamic zoom of an unmanned aerial vehicle according to claim 3, characterized in that: Establishing the real road surface coordinate system and the pixel imaging coordinate system and performing coordinate system projection mapping includes the following steps: The real-time projection point of the drone on the ground is taken as the origin, the forward direction of the drone is taken as the y-axis, and the The ground normal vector of the point is the z axis, the plane The normal vector of is the x-axis, and the real road coordinate system is established; The normal vector of the plane where the sensor at the center of the drone lens is located is Axis, the intersection of this axis and the plane where the sensor is located is the origin , the direction of sensor width is Axis, the direction where the sensor is high is Axis establishes pixel imaging coordinate system; Through the equation , Perform coordinate system projection mapping, where ( ) is the coordinate of the target in the pixel imaging coordinate system, is the coordinate of the target in the real road coordinate system, is the maximum image aspect ratio of the sensor, is the pitch angle of the drone, is the image pixel width, is the camera perspective, The flight altitude of the drone; Through the equation Get the target speed at time t ,in is the position of the UAV at time t in the real road coordinate system, is the position of the UAV at time t-1 in the real road coordinate system, and the flight speed scalar of the UAV at time t is , the video frame rate is f.

5. The method for measuring speed based on dynamic zoom of a drone according to claim 4, characterized in that: Through the equation Suppress the random noise generated by the jitter during the flight of the drone and obtain the suppressed speed of the target at time t ,in is the relative coordinate vector of the target to the UAV at time t obtained by the coordinate transformation algorithm, is the relative coordinate vector of the target to the UAV at time tk obtained by the coordinate transformation algorithm, and the flight speed vector of the UAV at time ti is , k is the time interval, k is 10.

6. The method for measuring speed based on dynamic zoom of a drone according to claim 4, characterized in that: The drone dynamically adjusts the focal length parameters through optical zoom and digital zoom. When it is digital zoom, the detection frame coordinates are scaled according to the zoom ratio; when it is optical zoom, the equivalent focal length parameters of the lens are updated.

7. The method for measuring speed based on dynamic zoom of a UAV according to claim 6, characterized in that: When digital zoom is used, the equation Add the center point position transformation to calibrate the target speed calculation, where is the zoom magnification, The minimum magnification of different lenses.

8. A patrol system for speed measurement based on dynamic zoom of a drone, using the speed measurement method based on dynamic zoom of a drone as claimed in claim 7, characterized in that: include: Monitoring module, used to monitor targets on the road through drones and obtain video streams; The target classification and recognition module processes the video stream to identify the types of targets, including cars, buses, trucks, and pedestrians; The road congestion detection module receives the video stream and the target type to calculate the number of cars, buses, and trucks, and then determines whether the road is congested and outputs the result; The vehicle speed calculation module receives the video stream, road congestion conditions, and target types to calculate the vehicle speed. When the road is not congested, it marks and alerts vehicles that are below the speed limit. The pedestrian detection module is used to receive video streams and target types for pedestrian detection, and mark and warn when pedestrians are detected on the road.

9. An electronic device comprising: at least one processor; as well as a memory communicatively connected to at least one processor; wherein, The memory stores instructions that can be executed by at least one processor. The instructions are executed by the at least one processor so that the at least one processor can perform the method according to any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images