Method and device for determining longitude and latitude of object based on unmanned aerial vehicle

Abstract

The invention provides a method and a device for determining longitude and latitude of an object based on an unmanned aerial vehicle. The method comprises the following steps: shooting an area needing to be inspected by using a pan-tilt camera of the unmanned aerial vehicle; identifying an object needing to be positioned from the shot plane image; determining the height of the object; identifying topographic features of the object from the plane image; calculating the height difference of the object relative to the unmanned aerial vehicle according to the height of the unmanned aerial vehicle relative to the ground, the height of the object and the height of the topographic features; calculating the horizontal distance and azimuth angle of the object relative to the unmanned aerial vehicle according to the height difference of the object relative to the unmanned aerial vehicle, the attitude parameters of the unmanned aerial vehicle and the attitude parameters of the pan-tilt camera; and calculating the longitude and latitude of the object according to the horizontal distance and azimuth angle of the object relative to the unmanned aerial vehicle and the longitude and latitude of the unmanned aerial vehicle. According to the scheme, the longitude and latitude of the object can be accurately calculated based on the longitude and latitude of the unmanned aerial vehicle. Even if the distance between the unmanned aerial vehicle and the object is long, the longitude and latitude of the object can be accurately calculated.

Description

Technical Field

[0001] The present invention relates to the field of drone inspection technology, and in particular to a method and device for determining the latitude and longitude of an object based on a drone. Background Art

[0002] During drone security inspections, locating the location of identified objects provides precise geographic information for subsequent intervention. Longitude and latitude coordinates are often used. For example, when a drone's gimbal camera detects a person, vehicle, or faulty equipment, the longitude and latitude coordinates can be used to quickly dispatch security personnel, patrol vehicles, or maintenance engineers to the scene to address the situation.

[0003] Currently, the longitude and latitude determined by the drone's GPS (Global Positioning System) positioning are typically used as the longitude and latitude of the identified object. When the drone is close to the identified object, the difference between their longitude and latitude is small, so the drone's longitude and latitude can be used as the object's longitude and latitude. However, because a drone's field of view can reach several hundred meters, when the distance between the drone and the object is greater, the longitude and latitude of the drone and the object differ significantly. If the drone's longitude and latitude are still used as the object's longitude and latitude, the object's location will be inaccurate, which will affect the timeliness of on-site handling and the efficiency of personnel. Summary of the Invention

[0004] The present invention provides a method and device for determining the latitude and longitude of an object based on a drone, so as to solve the technical problem of how to accurately calculate the latitude and longitude of an object.

[0005] To solve the above technical problems, the present invention provides a method for determining the latitude and longitude of an object based on a drone, comprising the following steps:

[0006] S1. Use the drone's gimbal camera to photograph the area that needs to be inspected;

[0007] S2, identifying the object to be located from the captured plane image;

[0008] S3. Determine the height of the object;

[0009] S4. Identifying the terrain features where the object is located from the planar image;

[0010] S5. Calculating a height difference between the object and the drone based on the height of the drone relative to the ground, the height of the object, and the height of the terrain feature;

[0011] S6, calculating a horizontal distance and an azimuth angle of the object relative to the UAV according to the height difference of the object relative to the UAV, the attitude parameter of the UAV and the attitude parameter of the gimbal camera;

[0012] S7, calculating the latitude and longitude of the object according to the horizontal distance and the azimuth angle of the object relative to the UAV and the latitude and longitude of the UAV.

[0013] Preferably, the step S3 comprises the following steps: judging the category of the object, and determining the height of the object according to the category of the object and a preset mapping relationship table, wherein the mapping relationship table stores heights corresponding to a plurality of objects respectively.

[0014] Preferably, after the step of determining the height of the object, the method further comprises the following steps: judging the attitude of the object, and correcting the height of the object according to the attitude of the object.

[0015] Preferably, the step S5 comprises the following steps: obtaining the height h1 of the UAV relative to the ground and the height h2 of the terrain feature; and calculating the height difference H of the object relative to the UAV according to the formula H = h1-(h2+h3), wherein h3 represents the height of the object.

[0016] Preferably, the step of calculating the horizontal distance of the object relative to the UAV in the step S6 comprises the following steps: obtaining an angle A1 of a vertical field of view width of the gimbal camera and a pixel coordinate y1 of the object in a height direction in the planar image;

[0017] calculating an angle a1 between a line connecting the gimbal camera and the object and a center line of the gimbal camera according to the formula a1 = R1 / A1 / (y1-R1*0.5), wherein R1 represents a resolution of the planar image in the vertical direction;

[0018] obtaining an included angle a2 between the center line of the gimbal camera and a vertical direction;

[0019] calculating an included angle a3 between the line connecting the gimbal camera and the object and the vertical direction according to the formula a3 = a1+a2;

[0020] calculating the horizontal distance L of the object relative to the UAV according to the formula L = H / tan(a3).

[0021] Preferably, the step of calculating the azimuth angle of the object relative to the UAV in the step S6 comprises the following steps: obtaining an angle A2 of a horizontal field of view width of the gimbal camera and a horizontal pixel coordinate x1 of a center of the object in the planar image;

[0022] According to a formula s1=R2 / A2 / (x1-R2*0.5), an included angle s1 between a line connecting a center of the object and the gimbal camera and a center line of the gimbal camera is calculated; wherein R2 represents a resolution of the planar image in a horizontal direction;

[0023] An included angle s2 between the center line of the gimbal camera and a nose direction of the unmanned aerial vehicle and an included angle s3 between the nose direction of the unmanned aerial vehicle and a north direction are obtained;

[0024] According to a formula S=s1+s2+s3, an azimuth angle S of the object relative to the unmanned aerial vehicle is calculated.

[0025] Preferably, the step S7 comprises the following steps: obtaining a current latitude N0 and a current longitude E0 of the unmanned aerial vehicle;

[0026] According to a formula d1=L*cos(S), a north component d1 of the object relative to the unmanned aerial vehicle is calculated;

[0027] According to a formula d2=L*sin(S), an east component d2 of the object relative to the unmanned aerial vehicle is calculated;

[0028] According to a formula d3=d1 / 111320, a latitude change d3 is calculated;

[0029] According to a formula d4=d2 / (111320*cos(N0)), a longitude change d4 is calculated;

[0030] According to a formula N1=N0+d3, a latitude N1 of the object is calculated;

[0031] According to a formula E1=E0+d4, a longitude E1 of the object is calculated.

[0032] The application further provides a device for determining latitude and longitude of an object based on an unmanned aerial vehicle, comprising the following modules:

[0033] A control module is configured to use a gimbal camera of the unmanned aerial vehicle to shoot an area needing to be inspected;

[0034] An object recognition module is configured to recognize an object needing to be positioned from a planar image shot;

[0035] A height determination module is configured to determine a height of the object;

[0036] A terrain feature recognition module is configured to recognize a terrain feature where the object is located from the planar image;

[0037] A calculation module is configured to calculate a height difference of the object relative to the unmanned aerial vehicle according to a height of the unmanned aerial vehicle relative to the ground, the height of the object and a height of the terrain feature.

[0038] The calculation module is further configured to calculate the horizontal distance and azimuth angle of the object relative to the drone based on the height difference of the object relative to the drone, the attitude parameters of the drone, and the attitude parameters of the gimbal camera;

[0039] The calculation module is further used to calculate the longitude and latitude of the object based on the horizontal distance and azimuth angle of the object relative to the drone and the longitude and latitude of the drone.

[0040] Preferably, the height determination module is used to perform the following steps: determine the category of the object, and determine the height of the object based on the category of the object and a preset mapping relationship table; wherein the mapping relationship table stores heights corresponding to multiple objects.

[0041] Preferably, the height determination module is further configured to perform the following steps: determining the posture of the object, and correcting the height of the object according to the posture of the object.

[0042] The present invention provides a method and device for determining the longitude and latitude of an object using a drone. The method first calculates the height difference of the object relative to the drone based on the drone's height relative to the ground, the object's height, and the height of terrain features. The method then calculates the horizontal distance and azimuth angle of the object relative to the drone based on the height difference, the drone's attitude parameters, and the attitude parameters of a gimbal camera. Finally, the object's longitude and latitude are calculated based on the object's horizontal distance and azimuth angle relative to the drone, as well as the drone's longitude and latitude. This method accurately calculates the object's longitude and latitude based on the drone's longitude and latitude. This method allows accurate calculation of the object's longitude and latitude, even when the drone is far from the object. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The present invention provides a flowchart of a method for determining the longitude and latitude of an object based on a drone, according to an embodiment of the present invention.

[0044] Figure 2 This is a schematic diagram of identifying a car using an object recognition model provided by an embodiment of the present invention.

[0045] Figure 3 This is a schematic diagram of identifying the terrain features where a car is located using a terrain feature recognition model provided by an embodiment of the present invention.

[0046] Figure 4 This is a relationship diagram of the height difference and horizontal distance between an identified vehicle and a drone, provided by an embodiment of the present invention.

[0047] Figure 5 This is a diagram showing the azimuth angle relationship between an identified vehicle and a drone, according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the objects, advantages and features of the present application clearer, a method and device for determining the latitude and longitude of an object based on a UAV is described in further detail below with reference to the drawings. It should be noted that the drawings are very simplified and use non-precise proportions, only to facilitate and clarify the purpose of illustrating the embodiments of the present application.

[0049] In the description of the present application, the terms "first", "second", and the like qualifiers are added for the convenience of description and reference, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features qualified with "first", "second", and the like can explicitly or implicitly include one or more of the features.

[0050] As shown in Figure 1 The present embodiment provides a method for determining the latitude and longitude of an object based on a UAV, comprising the following steps:

[0051] S1, using a gimbal camera of a UAV to shoot an area that needs to be inspected. The UAV can be flown to a certain height, and then the gimbal camera carried on the UAV is used to shoot the area that needs to be inspected. For example, the gimbal camera can shoot a photo with a resolution of 3000x4000, wherein the resolution R1 of the photo, i.e. the planar image, in the vertical direction is 3000 pixels, and the resolution R2 in the horizontal direction is 4000 pixels.

[0052] S2, identifying the object that needs to be positioned from the shot planar image. As shown in Figure 2 Yolo-v9 (You Only Look Once version 9) and RetinaNet (Retina Network) object detection models can be used to identify the object that needs to be positioned, Figure 2 The red box in the figure indicates the identification frame, and the current identified object is a car.

[0053] S3, determining the height of the object. Preferably, step S3 includes the following steps: judging the category of the object, and determining the height of the object according to the category of the object and a preset mapping relationship table. The mapping relationship table stores the heights corresponding to a plurality of objects respectively. After the step of determining the height of the object, the following steps can also be included: judging the posture of the object, and correcting the height of the object according to the posture of the object. For example, if the identified object is a person, the mapping relationship table stores that the height of the person is 1.7 meters, and at this time, it can be determined that the height of the person in the planar image is 1.7 meters. Then, it is judged which one of standing, squatting and lying is the current state of the person, if standing, then 1.7 meters, if squatting, then 1 meter, and if lying, then 0.5 meters. The height of the upper edge of the recognition frame can be taken as the height of the person. If the identified object is a small car, it can be judged whether the vehicle relative to the horizontal axis of the image is parallel, vertical or inclined, if parallel, then the height of the center of the recognition frame is taken as 1 meter, if vertical, then 1.5 meters, and if inclined, then 1.25 meters, and the height of the center of the recognition frame can be taken as the height of the vehicle.

[0054] S4, identifying the terrain feature where the object is located from the planar image. A multimodal model such as Qwen2.5-VL (Qwen2.5-Vision-Language) model of Tong Yiqian can be used to judge whether there is a viaduct or a bridge hole and other terrain features in the identified photo, and if so, whether the object such as a person or a vehicle is on the viaduct or in the bridge hole. Referring to Figure 3 If it is a general viaduct or a two-story platform, the height of the object can be increased by 5 meters, if it is at the entrance of a tunnel, the height of the object can be reduced by 5 meters, and if it is in some area with clear horizontal height change, more accurate data can be used to adjust the height of the object. Figure 3 The green coverage area in FIG. 1 shows that the identified terrain feature is a two-story platform. If a multimodal large model is not used, a special visual small model capable of identifying viaducts, bridge holes or specific terrain changes can also be used.

[0055] S5, calculating the height difference of the object relative to the unmanned aerial vehicle according to the height of the unmanned aerial vehicle relative to the ground, the height of the object and the height of the terrain feature. Preferably, referring to Figure 4As shown, step S5 comprises the following steps: obtaining the height h1 of the UAV relative to the ground and the height h2 of the terrain feature; calculating the height difference H of the object relative to the UAV according to the formula H = h1-(h2+h3); wherein h3 represents the height of the object. The height h1 of the UAV relative to the ground can be obtained by a GPS (Global Positioning System) or a height sensor on the UAV, and the height h2 of the terrain feature can be obtained by a preset mapping relationship table which can store the heights corresponding to different terrain features. In other embodiments, the result calculated according to the formula H = h1-(h2+h3) can be verified, and the theoretical value H can be corrected according to the verification result.

[0056] S6, according to the height difference of the object relative to the UAV, the attitude parameter of the UAV and the attitude parameter of the gimbal camera, calculating the horizontal distance and azimuth angle of the object relative to the UAV. Preferably, referring to Figure 4 As shown, the step of calculating the horizontal distance of the object relative to the UAV in step S6 comprises the following steps: obtaining the angle A1 of the vertical field of view width of the gimbal camera and the pixel coordinate y1 of the object in the height direction in the planar image; calculating the angle a1 between the line connecting the gimbal camera and the object and the median line z of the gimbal camera according to the formula a1 = R1 / A1 / (y1-R1*0.5); wherein R1 represents the resolution of the planar image in the vertical direction; obtaining the included angle a2 between the median line z of the gimbal camera and the vertical direction; calculating the included angle a3 between the line connecting the gimbal camera and the object and the vertical direction according to the formula a3 = a1+a2; calculating the horizontal distance L of the object relative to the UAV according to the formula L = H / tan(a3). R1 and a2 are the shooting parameters of the gimbal camera, which can be obtained by the gimbal camera. The angle A1 of the vertical field of view width of the gimbal camera is a parameter of the gimbal camera itself, which can be directly obtained from the gimbal camera; or the parameters of the gimbal camera itself are queried and stored in advance for subsequent direct retrieval. The entire planar image can be regarded as a coordinate system, for example, the lower left corner of the planar image is regarded as the origin, the bottom edge of the planar image is regarded as the x-axis, and the left vertical edge of the planar image is regarded as the y-axis. When the object is a person, the height coordinate of the upper edge of the recognition frame can be regarded as the pixel coordinate y1 of the person in the height direction in the planar image. When the object is a car, the height coordinate of the center of the recognition frame can be regarded as the pixel coordinate y1 of the car in the height direction in the planar image.

[0057] Preferably, referring to Figure 5As shown, the step of calculating the azimuth angle of the object relative to the UAV in step S6 comprises the following steps: obtaining an angle A2 of a horizontal field of view width of the gimbal camera and horizontal pixel coordinates x1 of the center of the object in the planar image; calculating an included angle s1 between a line connecting the center of the object and the gimbal camera and a center line z of the gimbal camera according to a formula s1 = R2 / A2 / (x1-R2*0.5); wherein R2 represents a resolution of the planar image in the horizontal direction; obtaining an included angle s2 between the center line z of the gimbal camera and a nose direction of the UAV, and an included angle s3 between the nose direction of the UAV and the true north direction; and calculating the azimuth angle S of the object relative to the UAV according to a formula S = s1 + s2 + s3. R2 and s2 are shooting parameters of the gimbal camera and can be obtained by the gimbal camera. s3 is an operating parameter of the UAV and can be obtained by the UAV. The angle A2 of the horizontal field of view width of the gimbal camera is a parameter of the gimbal camera itself and can be directly obtained from the gimbal camera; or the parameter of the gimbal camera itself is queried and stored in advance for subsequent direct retrieval. The horizontal pixel coordinates of the center of the recognition frame can be taken as the horizontal pixel coordinates x1 of the object. In other embodiments, the azimuth angle of the object relative to the UAV can also be calculated according to an included angle between the nose direction of the UAV and the true south direction.

[0058] S7, calculating the latitude and longitude of the object according to the horizontal distance and the azimuth angle of the object relative to the UAV and the latitude and longitude of the UAV. Preferably, step S7 comprises the following steps: obtaining the current latitude N0 and longitude E0 of the UAV; calculating a north component d1 of the object relative to the UAV according to a formula d1 = L*cos(S); calculating an east component d2 of the object relative to the UAV according to a formula d2 = L*sin(S); calculating a latitude change d3 according to a formula d3 = d1 / 111320; calculating a longitude change d4 according to a formula d4 = d2 / (111320*cos(N0)); calculating the latitude N1 of the object according to a formula N1 = N0+d3; and calculating the longitude E1 of the object according to a formula E1 = E0+d4. The current latitude and longitude of the UAV can be obtained by the GPS on the UAV.

[0059] The embodiment provides a method for determining longitude and latitude of an object based on a UAV, which comprises the following steps: calculating a height difference of the object relative to the UAV according to a height of the UAV relative to the ground, a height of the object and a height of a terrain feature; calculating a horizontal distance and an azimuth angle of the object relative to the UAV according to the height difference of the object relative to the UAV, an attitude parameter of the UAV and an attitude parameter of a gimbal camera; and calculating longitude and latitude of the object according to the horizontal distance and the azimuth angle of the object relative to the UAV and longitude and latitude of the UAV, so that the longitude and latitude of the object can be accurately calculated based on the longitude and latitude of the UAV. Even if the distance between the UAV and the object is far, the longitude and latitude of the object can be accurately calculated.

[0060] Based on the same technical concept as the method for determining longitude and latitude of an object based on a UAV, the embodiment provides a device for determining longitude and latitude of an object based on a UAV, which comprises the following modules.

[0061] A control module is configured to use a gimbal camera of the UAV to shoot an area needing to be inspected;

[0062] An object recognition module is configured to recognize an object needing to be positioned from a planar image shot;

[0063] A height determination module is configured to determine a height of the object;

[0064] A terrain feature recognition module is configured to recognize a terrain feature where the object is located from the planar image;

[0065] A calculation module is configured to calculate a height difference of the object relative to the UAV according to a height of the UAV relative to the ground, a height of the object and a height of the terrain feature;

[0066] The calculation module is further configured to calculate a horizontal distance and an azimuth angle of the object relative to the UAV according to the height difference of the object relative to the UAV, an attitude parameter of the UAV and an attitude parameter of the gimbal camera;

[0067] The calculation module is further configured to calculate longitude and latitude of the object according to the horizontal distance and the azimuth angle of the object relative to the UAV and longitude and latitude of the UAV.

[0068] The device for determining the longitude and latitude of an object based on a UAV provided by the embodiment calculates the height difference of the object relative to the UAV according to the height of the UAV relative to the ground, the height of the object and the height of the terrain feature, calculates the horizontal distance and the azimuth angle of the object relative to the UAV according to the height difference of the object relative to the UAV, the attitude parameter of the UAV and the attitude parameter of the gimbal camera, and finally calculates the longitude and latitude of the object according to the horizontal distance and the azimuth angle of the object relative to the UAV and the longitude and latitude of the UAV, so that the longitude and latitude of the object can be accurately calculated based on the longitude and latitude of the UAV. Even if the distance between the UAV and the object is far, the longitude and latitude of the object can also be accurately calculated.

[0069] Preferably, the height determining module is configured to determine the height of the object according to the category of the object and a preset mapping relationship table, wherein the mapping relationship table stores the heights corresponding to the plurality of objects respectively. The heights corresponding to different objects are stored in advance through the preset mapping relationship table, so that the convenience and accuracy of calculating the height of the object are improved.

[0070] Preferably, the height determining module is further configured to correct the height of the object according to the attitude of the object. The height of the object is corrected through the attitude of the object, so that the accuracy of the calculation result is improved.

[0071] In summary, the method and device for determining the longitude and latitude of an object based on a UAV provided by the embodiment calculate the height difference of the object relative to the UAV according to the height of the UAV relative to the ground, the height of the object and the height of the terrain feature, calculate the horizontal distance and the azimuth angle of the object relative to the UAV according to the height difference of the object relative to the UAV, the attitude parameter of the UAV and the attitude parameter of the gimbal camera, and finally calculate the longitude and latitude of the object according to the horizontal distance and the azimuth angle of the object relative to the UAV and the longitude and latitude of the UAV, so that the longitude and latitude of the object can be accurately calculated based on the longitude and latitude of the UAV. Even if the distance between the UAV and the object is far, the longitude and latitude of the object can also be accurately calculated.

[0072] The above description is only a description of the preferred embodiments of the present application, and does not limit the scope of the present application in any way. Any modification or modification of the present application made by a person of ordinary skill in the art based on the above disclosure is within the scope of the present application.

Claims

1. A method for determining the latitude and longitude of an object based on a drone, characterized in that: The following steps are involved: S1. Use the drone's gimbal camera to photograph the area that needs to be inspected; S2, identifying the object to be located from the captured plane image; S3, determining the height of the object; S4. Identifying the terrain features where the object is located from the planar image; S5. Calculating a height difference between the object and the drone based on the height of the drone relative to the ground, the height of the object, and the height of the terrain feature; S6. Calculate the horizontal distance and azimuth angle of the object relative to the drone based on the height difference of the object relative to the drone, the attitude parameters of the drone, and the attitude parameters of the gimbal camera; S7. Calculate the longitude and latitude of the object based on the horizontal distance and azimuth angle of the object relative to the drone and the longitude and latitude of the drone.

2. The method for determining the latitude and longitude of an object based on a drone according to claim 1, wherein: Step S3 includes the following steps: judging the category of the object, and determining the height of the object according to the category of the object and a preset mapping relationship table; wherein the mapping relationship table stores heights corresponding to a plurality of objects.

3. The method for determining the latitude and longitude of an object based on a drone as claimed in claim 2, wherein: After the step of determining the height of the object, the method further includes the following steps: judging the posture of the object and correcting the height of the object according to the posture of the object.

4. The method for determining the longitude and latitude of an object based on a drone according to claim 1, wherein: Step S5 includes the following steps: obtaining the height h1 of the drone relative to the ground and the height h2 of the terrain feature; calculating the height difference H of the object relative to the drone according to the formula H=h1-(h2+h3); wherein h3 represents the height of the object.

5. The method for determining the latitude and longitude of an object based on a drone as claimed in claim 4, wherein: The step of calculating the horizontal distance of the object relative to the drone in step S6 includes the following steps: obtaining an angle A1 of the vertical field of view width of the gimbal camera and a pixel coordinate y1 of the height direction of the object in the plane image; Calculate the angle a1 between the line connecting the gimbal camera and the object and the midline of the gimbal camera according to the formula a1=R1 / A1 / (y1-R1×0.5); where R1 represents the vertical resolution of the planar image; Obtaining the angle a2 between the center line of the gimbal camera and the vertical direction; Calculate the angle a3 between the vertical direction and the line connecting the gimbal camera and the object according to the formula a3=a1+a2; According to the formula L=H / tan(a3), the horizontal distance L of the object relative to the drone is calculated.

6. The method for determining the latitude and longitude of an object based on a drone as claimed in claim 5, wherein: The step of calculating the azimuth angle of the object relative to the drone in step S6 comprises the following steps: obtaining the angle A2 of the horizontal field of view width of the gimbal camera and the horizontal pixel coordinate x1 of the center of the object in the plane image; Calculate the angle s1 between the line connecting the center of the object and the gimbal camera and the midline of the gimbal camera according to the formula s1 = R2 / A2 / (x1-R2×0.5); where R2 represents the horizontal resolution of the planar image; Obtaining the angle s2 between the center line of the gimbal camera and the nose direction of the UAV, and the angle s3 between the nose direction of the UAV and the true north direction; According to the formula S=s1+s2+s3, the azimuth angle S of the object relative to the drone is calculated.

7. The method for determining the latitude and longitude of an object based on a drone as claimed in claim 6, wherein: Step S7 includes the following steps: obtaining the current latitude N0 and longitude E0 of the UAV; Calculate the north component d1 of the object relative to the drone according to the formula d1=L×cos(S); Calculate the east component d2 of the object relative to the drone according to the formula d2=L×sin(S); Calculate the latitude change d3 using the formula d3 = d1 / 111320; Calculate the longitude change d4 using the formula d4 = d2 / (111320 × cos(N0)); Calculate the latitude N1 of the object according to the formula N1=N0+d3; According to the formula E1=E0+d4, the longitude E1 of the object is calculated.

8. A device for determining the latitude and longitude of an object based on a drone, characterized in that: Includes the following modules: A control module is used to use the drone's pan / tilt camera to capture images of the area to be inspected; An object recognition module is used to identify objects that need to be located from the captured plane image; a height determination module, configured to determine the height of the object; A terrain feature recognition module, configured to recognize the terrain feature where the object is located from the planar image; a calculation module, configured to calculate a height difference of the object relative to the drone based on the height of the drone relative to the ground, the height of the object, and the height of the terrain feature; The calculation module is further configured to calculate the horizontal distance and azimuth angle of the object relative to the drone based on the height difference of the object relative to the drone, the attitude parameters of the drone, and the attitude parameters of the gimbal camera; The calculation module is further used to calculate the longitude and latitude of the object based on the horizontal distance and azimuth angle of the object relative to the drone and the longitude and latitude of the drone.

9. The device for determining the latitude and longitude of an object based on a drone as claimed in claim 8, characterized in that: The height determination module is used to perform the following steps: determine the category of the object, and determine the height of the object according to the category of the object and a preset mapping relationship table; wherein the mapping relationship table stores heights corresponding to multiple objects.

10. The device for determining the latitude and longitude of an object based on a drone as claimed in claim 9, characterized in that: The height determination module is further configured to perform the following steps: determining the posture of the object, and correcting the height of the object according to the posture of the object.

Images