Method and system for accurately measuring and calculating geographic coordinates of hidden danger target based on unmanned aerial vehicle and artificial intelligence detection, and computer readable storage medium

Abstract

The invention belongs to the technical field of intelligent hidden danger detection, and more specifically relates to a hidden danger target geographic coordinate accurate calculation method and system based on an unmanned aerial vehicle and artificial intelligence detection, and a computer readable storage medium. The method comprises the following steps: the unmanned aerial vehicle identifies a target in a video image in real time and analyzes a video frame; when analyzing that the current frame image contains the hidden danger target, the control interface notifies the unmanned aerial vehicle to hover; according to the pixel coordinates of the hidden danger target in the video image and the camera focal length, the camera negative film width and height and the imaging size of the unmanned aerial vehicle cradle head, calculating a cradle head relative adjustment angle which enables the hidden danger target to be centered; according to the obtained relative adjustment angle of the holder, adjusting the holder to enable the hidden danger target to be in the center of a holder picture; and after the adjustment is completed, calculating the GPS data of the hidden danger target. The problems that in the prior art, data processing lags behind, hidden dangers cannot achieve closed-loop management, system fragmentation lacks end-to-end collaborative optimization, and consequently the overall efficiency is limited are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent hidden danger detection, and more specifically, relates to a method, system and computer-readable storage medium for accurately calculating the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection. Background Art

[0002] Currently, drone technology and edge computing technology are developing rapidly. The gimbal pods carried by drones have realized the integration of high-precision laser ranging modules, high-performance telephoto cameras, etc., and edge computing terminals are highly integrated, which provides unlimited space for the application of new technologies on drones.

[0003] Chinese invention patent CN111612728A discloses a 3D point cloud densification method and device based on binocular RGB images. Step 1: Generate a depth image from the binocular RGB image; Estimate the approximate three-dimensional coordinates of each pixel in the depth image in the LiDAR coordinate system based on the depth information of the depth image; Step 2: Use the cyclic RANSAC algorithm to perform ground segmentation of the point cloud and extract the non-ground point cloud; Step 3: Insert the extracted non-ground point cloud into a KDTree, search for a predetermined number of neighboring points in the KDTree based on the approximate three-dimensional coordinates of each pixel in the LiDAR coordinate system, and use the neighboring points to perform surface reconstruction; Step 4: Based on the surface reconstruction result and the calibration parameters of the LiDAR and camera, derive the precise coordinates of the approximate three-dimensional coordinates by a computational geometry method, and fuse the precise coordinates with the original LiDAR point cloud to obtain a densified point cloud.

[0004] Traditional hidden danger target detection mostly relies on manual inspections or fixed camera monitoring, which has problems such as low efficiency, high cost, and limited coverage. In recent years, drone technology has gradually been applied to the field of hidden danger inspection due to its strong flexibility and rapid deployment. However, the existing technology still has the following pain points. First, data processing lags. The images / videos collected by drones rely on manual analysis at the back end, which lacks real-time performance and is difficult to meet the needs of emergency scenarios. Second, hidden dangers cannot be managed in a closed loop. Current technology only stops at being able to detect hidden dangers, and cannot obtain the actual coordinates of the hidden dangers. Further actions cannot be taken to achieve closed-loop processing of hidden danger targets. Third, the system is fragmented. UAV control, AI algorithms, and geographic information processing modules operate independently, lacking end-to-end collaborative optimization, resulting in limited overall efficiency. Summary of the Invention

[0005] The present invention aims to overcome at least one of the defects of the above-mentioned prior art and provide a method for accurately measuring the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection, which solves the problems of data processing lag, reliance on back-end manual analysis of images / videos collected by drones, and insufficient real-time performance; as well as the inability to manage hidden dangers in a closed loop. Current technology is only able to detect hidden dangers but cannot obtain the actual coordinate locations of hidden dangers.

[0006] On the other hand, the present invention also includes a system for implementing a method for accurately calculating the geographical coordinates of hidden danger targets based on drone and artificial intelligence detection; And, a computer-readable storage medium that executes the above-mentioned method for accurately measuring the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection.

[0007] The detailed technical solutions of the present invention are as follows: A method for accurately calculating the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection, the method comprising: S1. The drone uses hidden danger target detection combined with target tracking algorithm to identify targets in video images in real time and analyze video frames; S2. When the current frame image is analyzed to contain a hidden danger target, the onboard microcomputer immediately notifies the drone to hover through the drone's open control interface; S3. Calculate the relative adjustment angle of the gimbal to center the hidden danger target based on the pixel coordinates of the hidden danger target in the video image, the focal length of the drone gimbal camera, the width and height of the camera film, and the image size. The calculation formula is as follows: (1); (2); (3); (4); In formulas (1)~(4), and The center point of the hidden danger target detection box in the pixel coordinate system Relative to the center of the video image The two-axis pixel value deviation; The width of the drone video image, in pixels; The height of the drone video image, in pixels; are the width, height, coordinates of the upper left point on the u-axis, and the coordinates on the v-axis of the hidden danger target detection frame in the pixel coordinate system. The u-axis and v-axis are the horizontal and vertical coordinates of the pixel coordinate system respectively; 、 are the calculated relative adjustment angles of the UAV gimbal yaw and pitch, is the width of the camera film, is the camera film height, f is the focal length of the pan / tilt lens, in millimeters; S4. Based on the relative yaw adjustment angle and pitch adjustment angle of the drone gimbal obtained in step S3, notify the drone open interface to adjust the gimbal so that the hidden danger target is in the center of the gimbal image; after the adjustment is completed, calculate the GPS data of the hidden danger target, specifically including: S41: If the drone gimbal is integrated with a laser ranging module and the gimbal three-axis angles in a standard format are directly obtained through the drone open data interface, the GPS data of the hidden danger target is calculated as: (5); (6); (7); In formulas (5)~(7), 、 The pitch angle and yaw angle of the drone gimbal are directly obtained in radians; the laser ranging data of the drone is recorded as , in meters; 、 、 To obtain the latitude, longitude and altitude data of the drone, 、 、 To predict the latitude, longitude and altitude data of the hidden danger target, 、 、 、 The unit is degree, and The unit is meter, is 3.1415926.

[0008] S42: If the laser ranging module is integrated in the UAV gimbal, and the three-axis angle of the gimbal cannot be directly obtained in the standard format through the UAV open data interface, but it meets the requirements of "obtaining the pitch angle and yaw angle of the UAV and recording them as 、 ; It is the angle between the projection of the nose direction on the horizontal plane and the north direction. The north direction is 0 degrees, and the angle increases clockwise. The obtained yaw angle of the gimbal is the angle relative to the horizontal plane, and the downward angle is negative; the obtained gimbal pitch angle is the angle relative to the horizontal plane where the nose is located, and the downward angle is negative. In this case, the calculation of the hidden danger target GPS data is: (8); (9); (10); (11); (12); In formulas (8)~(12), 、 are the pitch and yaw angles of the UAV gimbal obtained after conversion.

[0009] S43: If the UAV gimbal does not have the aforementioned laser ranging module integrated or does not support the acquisition of ranging data, and the gimbal three-axis angles in a standard format can be directly obtained through the UAV open data interface, the calculation of the hidden danger target GPS data is: (13); (14); (15); (16); S44: If the UAV gimbal does not integrate the above laser ranging module or does not support the acquisition of ranging data, and the UAV open data interface cannot directly obtain the standard format of the gimbal three-axis angle, but meets the "obtain the UAV pitch angle and yaw angle, the unit is radian, recorded as 、 ; It is the angle between the projection of the nose direction on the horizontal plane and the north direction. The north direction is 0 degrees, and the angle increases clockwise. The obtained gimbal yaw angle is the angle with the horizontal plane, and the downward angle is negative; the obtained gimbal pitch angle is the angle with respect to the horizontal plane where the drone's nose is located, and the downward angle is negative. In this case, the calculation formula for the hidden danger target GPS data is: (17); (18); (19); (20); (twenty one); (twenty two).

[0010] In another aspect of the present invention, a system for accurately calculating the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection is provided. The system includes a drone multi-source data real-time communication module, a real-time video AI analysis module, and a hidden danger target geographic coordinate solution module: The UAV multi-source data real-time communication module has the function of carrying a microcomputer on the UAV and using the open source interface provided by the UAV manufacturer to allow the microcomputer to communicate with the UAV to obtain and send real-time data.

[0011] The real-time video AI analysis module deploys a hidden danger target detection network and combines it with a target tracking algorithm to perform real-time identification of targets in video images.

[0012] The hidden danger target geographic coordinate solution module has the function of establishing a conversion relationship from pixel coordinates to geographic coordinates based on the above-mentioned drone data after the real-time video AI analysis module identifies a hidden danger, and further obtaining the geographic coordinates of the hidden danger.

[0013] In another aspect of the present invention, a computer-readable storage medium is provided, which stores executable instructions. When the instructions are executed, the machine executes the above-mentioned method for accurately measuring the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a method for accurately calculating the geographic coordinates of hidden danger targets based on drones and artificial intelligence detection. A method for accurately calculating the geographic coordinates of hidden danger targets based on drones and artificial intelligence detection, by using an edge computing terminal to carry a drone, with the help of high-precision laser ranging data and AI analysis of the pan-tilt video and combined with the drone body data, realizes the conversion of two-dimensional objects in video images to three-dimensional high-precision positioning coordinate data in space; further promotes the closed-loop management of "monitoring-early warning-disposal" in the drone inspection area. The present invention deeply integrates drone control, AI algorithms and geographic information processing modules, realizes end-to-end collaborative optimization, and improves the overall performance of the system; during the flight, the drone automatically adjusts the flight path and shooting parameters according to the terrain information provided by the geographic information processing module and the hidden danger information identified by the AI algorithm, thereby realizing the possibility of intelligent drone inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a method for accurately calculating the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection as described in the present invention.

[0016] Figure 2 This is an example diagram of the pixel coordinate system and the meaning of each formula symbol in Example 1 of the present invention.

[0017] Figure 3 This is an example diagram of the relationship between the pan / tilt adjustment angle and the pixel value deviation in Example 1 of the present invention.

[0018] Figure 4 It is a system module diagram for implementing this method in Example 1 of the present invention. DETAILED DESCRIPTION

[0019] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0020] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0021] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0022] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0023] Example 1 Ginseng Figure 1 This embodiment provides a method for accurately calculating the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection, the method comprising: S1. UAV real-time identification and analysis of video frames: The UAV uses hidden target detection combined with target tracking algorithms to identify targets in video images in real time and analyze video frames; Preferably, the hidden danger target detection can be implemented based on the YOLOv8 algorithm, and the target tracking algorithm can be implemented based on the DeepSort algorithm.

[0024] S2. Detecting a hidden danger target and causing the drone to hover: When the current frame image is analyzed to contain a hidden danger target, the onboard microcomputer immediately notifies the drone to hover through the drone's open control interface; Preferably, this embodiment takes the collection of GPS true value data of hidden danger targets, whose longitude and latitude are 118.085950294 and 36.827100718 respectively as an example; S3. Calculate the relative adjustment angle of the PTZ to center the hidden danger target: According to the pixel coordinates of the hidden danger target in the video image and the focal length of the drone gimbal camera, the width and height of the camera film, and the imaging size, the relative adjustment angle of the gimbal to make the hidden danger target centered is calculated, that is, the relative yaw adjustment angle and the relative pitch adjustment angle of the drone gimbal, such as Figure 2 As shown, the calculation formula is as follows: (1); (2); (3); (4); In formulas (1)~(4), and The center point p of the hidden danger target detection box in the pixel coordinate system is relative to the center point of the video image The two-axis pixel value deviation; The width of the drone video image, in pixels; The height of the drone video image, in pixels; are the width, height, coordinates of the upper left point on the u-axis, and the coordinates on the v-axis of the hidden danger target detection frame in the pixel coordinate system. The u-axis and v-axis are the horizontal and vertical coordinates of the pixel coordinate system respectively; The relationship between the gimbal adjustment angle and pixel value deviation is as follows: Figure 3 As shown, oxy is the video image coordinate system, the x and y axes are parallel to the u and v axes, - - - is the camera coordinate system, - - - is the world coordinate system, 、 are the calculated relative adjustment angles of the UAV gimbal yaw and pitch, is the width of the camera film, is the camera film height, and f is the focal length of the pan / tilt lens, both in millimeters.

[0025] S4. Calculate GPS data of hidden danger targets: According to the relative adjustment angle of the gimbal obtained in step S3, i.e., the relative yaw adjustment angle and pitch adjustment angle of the gimbal of the UAV, the open interface of the UAV is notified to adjust the gimbal so that the hidden danger target is in the center of the gimbal image; after the adjustment is completed, the pitch angle and yaw angle of the UAV, the pitch angle and yaw angle of the gimbal, the laser ranging data of the UAV and the latitude, longitude and altitude data of the UAV at this moment are obtained, and the GPS data of the hidden danger target is calculated based on the above data, specifically including: Step S41: If the laser ranging module is integrated into the UAV gimbal, and the gimbal three-axis angle in a standard format is directly obtained through the UAV open data interface, the GPS data of the hidden danger target is calculated as follows: (5); (6); (7); In formulas (5)~(7), 、 The pitch angle and yaw angle of the drone gimbal are directly obtained by the computing terminal on the drone, and the unit is radian; the laser ranging data of the drone is recorded as , in meters; 、 、 To obtain the latitude, longitude and altitude data of the drone, 、 、 To predict the latitude, longitude and altitude data of the hidden danger target, 、 、 、 The unit is degree, and The unit is meter, is 3.1415926.

[0026] Step S42: If the laser ranging module is integrated in the UAV gimbal, and the three-axis angle of the gimbal cannot be directly obtained in the standard format through the UAV open data interface, but it meets the requirements of "obtaining the pitch angle and yaw angle of the UAV and recording them as 、 ; It is the angle between the projection of the nose direction on the horizontal plane and the north direction. The north direction is 0 degrees, and the angle increases clockwise. The obtained yaw angle of the gimbal is the angle relative to the horizontal plane, and the downward angle is negative; the obtained gimbal pitch angle is the angle relative to the horizontal plane where the nose is located, and the downward angle is negative. In this case, the calculation of the hidden danger target GPS data is: (8); (9); (10); (11); (12); In formulas (8)~(12), 、 are the calculated pitch and yaw angles of the UAV gimbal.

[0027] Step S43: If the UAV gimbal does not have the laser ranging module integrated or does not support the acquisition of ranging data, and the gimbal three-axis angle in a standard format can be directly obtained through the UAV open data interface, the GPS data of the hidden danger target is calculated as follows: (13); (14); (15); (16); Step S44: If the UAV gimbal does not integrate the laser ranging module or does not support the acquisition of ranging data, and the UAV open data interface cannot directly obtain the standard format of the gimbal three-axis angle, but meets the "obtain the pitch angle and yaw angle of the UAV in radians, record it as 、 ; It is the angle between the projection of the nose direction on the horizontal plane and the north direction. The north direction is 0 degrees, and the angle increases clockwise. The obtained gimbal yaw angle is the angle with the horizontal plane, and the downward angle is negative; the obtained gimbal pitch angle is the angle with respect to the horizontal plane where the drone's nose is located, and the downward angle is negative. In this case, the calculation formula for the hidden danger target GPS data is: (17); (18); (19); (20); (twenty one); (twenty two); In formulas (17) to (22), the interpretation of the above symbols is consistent with the above. Except for the above examples, the above formulas are applicable to other cases, except that the positive and negative signs need to be judged according to the specific situation.

[0028] Furthermore, the real-time ranging is performed by the method described in this embodiment, and the experimental data are as follows: Specifically, the drone route is planned around the hidden danger target, and the drone flight altitude is used as a variable. Waypoints are planned at relative heights of 20m, 40m, 60m, and 75m from the ground, respectively. A total of 13 waypoints are planned. Each time the waypoint is reached, the present invention is triggered to perform real-time video recognition, thereby triggering the calculation of the target GPS geographic coordinates.

[0029] When the UAV flies to a relative height of 20m above the ground, the GPS data and errors of the hidden danger targets calculated by the present invention are shown in Table 1, with the error unit being meters: Table 1 GPS data and error of 20m relative height of UAV from the ground

[0030] When the UAV flies to a relative height of 40m above the ground, the GPS data and errors of the hidden danger targets calculated by the present invention are shown in Table 2, with the error unit being meters: Table 2 GPS data and error when the UAV is 40m above the ground

[0031] When the UAV flies to a relative height of 60m and 75m above the ground, the GPS data and errors of the hidden danger targets calculated by the present invention are shown in Table 3, with the error unit being meters: Table 3 GPS data and errors when the UAV is at a relative height of 60m and 75m above the ground

[0032] The above embodiments show that, without relying on the laser ranging module integrated with the gimbal, the average error of the target GPS calculated by the present invention is approximately 0.45861 meters.

[0033] After testing, the planar positioning error in the embodiment of the present invention is less than 0.5 meters, achieving decimeter-level positioning accuracy. This invention achieves real-time and accurate mapping of video images and geographic information on a UAV mobile platform, resolving the problem of traditional equipment being unable to perform closed-loop processing when discovering hidden dangers. It also solves the problem of independent operation of the UAV control, AI algorithm, and geographic information processing modules, achieving end-to-end collaboration among the three modules. Applied to the field of power line inspection, it can accurately locate the geographic coordinates of hidden dangers; in the field of emergency rescue, it can quickly calibrate the location of geological disaster points; and shorten the response time for hidden danger disposal.

[0034] Example 2 This embodiment provides a system for implementing a method for accurately calculating the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection, the system comprising: The system includes a real-time communication module for multi-source data from drones, a real-time video AI analysis module, and a module for calculating the geographic coordinates of hidden danger targets. Figure 4 As shown: The drone's multi-source data real-time communication module allows the drone's onboard microcomputer to communicate with the drone using an open-source interface provided by the drone manufacturer to acquire and transmit real-time data. For example, it can acquire real-time high-precision RTK positioning data, including longitude, latitude, and altitude; real-time IMU inertial measurement data, including the three-axis attitude angles of the drone and gimbal; real-time laser ranging data and 4K-level visible light video streams; and it can send real-time control commands for drone hovering and gimbal adjustment.

[0035] The real-time video AI analysis module deploys a hidden danger target detection network and combines it with a target tracking algorithm to perform real-time identification of targets in video images.

[0036] The hidden danger target geographic coordinate solution module has the function of establishing a conversion relationship from pixel coordinates to geographic coordinates based on the above-mentioned drone data after the real-time video AI analysis module identifies a hidden danger, and further obtaining the geographic coordinates of the hidden danger.

[0037] Example 3 This embodiment also provides a computer-readable storage medium storing executable instructions, which, when executed, enable the machine to execute the above-mentioned method for accurately measuring the geographic coordinates of hidden danger targets based on drone and artificial intelligence detection.

[0038] Specifically, a system or device equipped with a readable storage medium can be provided, on which software program codes that implement the functions of any of the above-mentioned embodiments are stored, and a computer or processor of the system or device can read and execute instructions stored in the readable storage medium.

[0039] In this case, the program code itself read from the computer-readable medium can realize the function of any one of the above-mentioned embodiments, and thus the computer-readable code and the computer-readable storage medium storing the computer-readable code constitute part of this specification.

[0040] Examples of readable storage media include floppy disks, hard disks, magneto-optical disks, optical disks (e.g., CD-ROMs, CD-Rs, CD-RWs, DVD-ROMs, DVD-RAMs, DVD-RWs, DVD-RWs), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, the program code may be downloaded from a server computer or a cloud via a communication network.

[0041] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0042] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or boxes.

[0043] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0044] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0045] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the technical solutions of the present invention, and are not intended to limit the specific implementation methods of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the claims of the present invention shall be included within the scope of protection of the claims of the present invention.

Claims

1. A method for accurately calculating the geographical coordinates of hidden danger targets based on drone and artificial intelligence detection, characterized in that: The method comprises: S1. The drone uses hidden danger target detection combined with target tracking algorithm to identify targets in video images in real time and analyze video frames; S2. When the current frame image is analyzed to contain a hidden danger target, the onboard microcomputer immediately notifies the drone to hover through the drone's open control interface; S3. Calculate the relative adjustment angle of the gimbal to center the hidden danger target, i.e., the relative yaw adjustment angle and the relative pitch adjustment angle of the gimbal, based on the pixel coordinates of the hidden danger target in the video image, the focal length of the gimbal, the width and height of the camera film, and the image size. S4. Based on the relative adjustment angle of the gimbal obtained in step S3, the UAV open interface is notified to adjust the gimbal so that the hidden danger target is in the center of the gimbal image; after the adjustment is completed, the GPS data of the hidden danger target is calculated.

2. The method for accurately calculating the geographical coordinates of hidden danger targets based on drone and artificial intelligence detection according to claim 1 is characterized in that: The calculation of the relative adjustment angle of the pan / tilt head so as to center the hidden danger target specifically includes: （1）； （2）； （3）； （4）； In formulas (1)~(4), and The center point of the hidden danger target detection box in the pixel coordinate system Relative to the center of the video image The two-axis pixel value deviation; The width of the drone video image, in pixels; The height of the drone video image, in pixels; are the width, height, coordinates of the upper left point on the u-axis, and coordinates on the v-axis of the hidden danger target detection box in the pixel coordinate system; 、 are the calculated relative adjustment angles of the UAV gimbal yaw and pitch, is the width of the camera film, is the camera film height, and f is the focal length of the pan / tilt lens, both in millimeters.

3. The method for accurately calculating the geographical coordinates of hidden danger targets based on drone and artificial intelligence detection according to claim 2 is characterized in that: After the adjustment is completed, the GPS data of the hidden danger target is calculated, including: If the laser ranging module is integrated into the drone gimbal, and the three-axis angle of the gimbal in a standard format is directly obtained through the drone's open data interface, the GPS data of the hidden danger target is calculated as follows: （5）； （6）； （7）； In formulas (5)~(7), 、 The pitch angle and yaw angle of the drone gimbal are obtained in radians; the laser ranging data of the drone is recorded as , in meters; 、 、 To obtain the latitude, longitude and altitude data of the drone, 、 、 To predict the latitude, longitude and altitude data of the hidden danger target, 、 、 、 The unit is degree, and The unit is meter, is 3.1415926.

4. The method for accurately calculating the geographical coordinates of hidden danger targets based on drone and artificial intelligence detection according to claim 2 is characterized in that: After the adjustment is completed, the GPS data of the hidden danger target is calculated, which also includes: If the laser ranging module is integrated in the UAV gimbal, and the three-axis angle of the gimbal cannot be directly obtained in the standard format through the UAV open data interface, but the pitch angle and yaw angle of the UAV are obtained, record them as 、 ; It is the angle between the projection of the nose direction on the horizontal plane and the north direction. The north direction is 0 degrees, and the angle increases clockwise. The obtained yaw angle of the gimbal is the angle relative to the horizontal plane, and the downward angle is negative; the obtained gimbal pitch angle is the angle relative to the horizontal plane where the nose is located, and the downward angle is negative. In this case, the calculation of the hidden danger target GPS data is: （8）； （9）； （10）； （11）； （12）； In formulas (8)~(12), 、 are the pitch and yaw angles of the UAV gimbal obtained after conversion.

5. The method for accurately calculating the geographical coordinates of hidden danger targets based on drone and artificial intelligence detection according to claim 2 is characterized in that: After the adjustment is completed, the GPS data of the hidden danger target is calculated, which also includes: If the UAV gimbal does not have the aforementioned laser ranging module integrated or does not support the acquisition of ranging data, and the gimbal three-axis angles in a standard format can be directly obtained through the UAV open data interface, the calculation of the hidden danger target GPS data is: （13）； （14）； （15）； （16）。 6. The method for accurately calculating the geographical coordinates of hidden danger targets based on drone and artificial intelligence detection according to claim 2 is characterized in that: After the adjustment is completed, the GPS data of the hidden danger target is calculated, which also includes: If the UAV gimbal does not integrate the above laser ranging module or does not support the acquisition of ranging data, and the UAV open data interface cannot directly obtain the standard format of the gimbal three-axis angle, but meets the "obtain the UAV's pitch angle and yaw angle, the unit is radian, recorded as 、 ; It is the angle between the projection of the nose direction on the horizontal plane and the north direction. The north direction is 0 degrees, and the angle increases clockwise. The obtained gimbal yaw angle is the angle with the horizontal plane, and the downward angle is negative; the obtained gimbal pitch angle is the angle with respect to the horizontal plane where the drone's nose is located, and the downward angle is negative. In this case, the calculation formula for the hidden danger target GPS data is: （17）； （18）； （19）； （20）； （21）； （22）。 7. A system for accurately calculating the geographical coordinates of hidden danger targets based on drone and artificial intelligence detection, characterized in that: The system includes a real-time communication module for multi-source data from drones, a real-time video AI analysis module, and a module for calculating the geographic coordinates of hidden danger targets: The UAV multi-source data real-time communication module is a microcomputer carried by the UAV, which communicates with the UAV through the open-source interface of the UAV manufacturer to obtain and send real-time data; The real-time video AI analysis module deploys a hidden danger target detection network and combines it with a target tracking algorithm to perform real-time recognition of targets in video images. The hidden danger target geographic coordinate solution module, when the real-time video AI analysis module identifies the hidden danger, establishes a conversion relationship from pixel coordinates to geographic coordinates based on the above-mentioned drone data, and finally obtains the geographic coordinates of the hidden danger.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images