Panoramic image-based power grid target positioning method, device, storage medium and system
Through the panoramic image target detection model and vector cross product method, the problem of automatic distinction and positioning of small and medium targets in panoramic images is solved, and efficient and low-cost multi-target positioning is achieved.
Patent Information
- Application Number
- CN202510898385.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-01
- Publication Date
- 2025-09-23
AI Technical Summary
Existing technologies have difficulty automatically distinguishing and accurately locating multiple small targets of the same type in panoramic images, such as transformer boxes and charging piles, especially in the presence of obstructions, and traditional methods are costly or inefficient.
The target detection model of panoramic images is used to identify small targets, construct the line of sight from the viewpoint to the target, eliminate false intersections through the vector cross product method, and calculate the coordinates of the real intersection points to achieve target positioning.
It realizes the automatic distinction and precise positioning of multiple small targets of the same type in panoramic images, improves positioning efficiency and reduces costs.
Smart Images
Figure CN120689579A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a method, device, storage medium and system for positioning a power grid target based on panoramic images, and belongs to the technical field of small target positioning in power grids. Background Art
[0002] Small objects on the power grid, such as transformers and charging stations, are crucial nodes in grid operation. The digitalization of power systems requires the precise location of these small objects. Measurement methods based on portable instruments are time-consuming, labor-intensive, and inefficient. Methods based on satellite remote sensing or drone imagery are difficult to detect due to the small size of the objects or impractical due to obstructions (such as buildings or trees). LiDAR point cloud data can accurately extract the geometric features of objects, but this is expensive and requires large amounts of data.
[0003] As a type of ground-based imagery, panoramic images offer the advantages of wide spatial coverage, rich recorded information, easy acquisition, and low cost. Small targets are clearly imaged in panoramic images, facilitating automated detection. In theory, as long as a target appears in two panoramic images simultaneously, a simulated line of sight can be constructed using the information recorded by the panoramic images, and the target position can be calculated using the forward intersection method. However, small targets such as substations and charging stations have similar appearances, making automatic differentiation and location difficult when multiple objects of the same type appear in two panoramic images. Summary of the Invention
[0004] The purpose of the present invention is to provide a method, device, storage medium and system for power grid target positioning based on panoramic images, so as to realize automatic differentiation and positioning of multiple targets of the same type.
[0005] To achieve the above objectives, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for positioning a power grid target based on panoramic images, comprising: After obtaining panoramic images from two adjacent viewpoints, they are input into the trained object detection model to identify all objects and mark them with detection boxes. The bottom midpoint of the detection box corresponding to the target is used as the target reference point to obtain the pixel coordinates of the target reference point. Based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point, a line of sight from the viewpoint to the target reference point is constructed, and the direction vector of the line of sight is calculated. All viewpoints and target reference points are polled to obtain all lines of sight. Combine the sight lines of two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersection points to obtain all remaining intersection points; The actual intersection points are selected from all the remaining intersection points, and the coordinates of the actual intersection points are calculated and used as the coordinates of the target to complete the power grid target positioning.
[0006] Furthermore, the panoramic image based on the viewpoint and the pixel coordinates of the target reference point is used to construct a line of sight from the viewpoint to the target reference point, and calculate the direction vector of the line of sight, including: Get the lens direction in the panoramic image of the viewpoint; Calculate the direction vector of the target reference point in the camera coordinate system; Converting the direction vector of the target reference point in the camera coordinate system into the direction vector of the target reference point in the global coordinate system based on the lens direction, and taking the direction corresponding to the direction vector of the target reference point in the global coordinate system as the line of sight from the viewpoint to the target reference point; The direction vector of the sight line is the direction vector of the target reference point in the global coordinate system; The camera coordinate system is a three-dimensional rectangular coordinate system established with the camera optical center as the origin. The Z axis coincides with the optical axis and points in the imaging direction. The X axis is parallel to the horizontal direction of the image plane, and the Y axis is parallel to the vertical direction of the image plane. Among them, the global coordinate system is the world coordinate system in three-dimensional space.
[0007] Furthermore, the sight lines of the two viewpoints are combined in pairs, and the sight line intersection analysis is performed by the vector cross product method, and the pseudo intersection points are eliminated to obtain all the remaining intersection points, including: Combine the sight lines of the two viewpoints in pairs to obtain all intersection points; Calculate the cross product of the direction vectors of the two combined sight lines; If the cross product is greater than 0 and the target is located on the right side of the travel direction, the intersection point corresponding to the cross product is a pseudo-intersection point, and the pseudo-intersection point is discarded; If the cross product is less than 0 and the target is located on the left side of the travel direction, the intersection point corresponding to the cross product is a pseudo-intersection point, and the pseudo-intersection point is discarded; After eliminating all pseudo intersections, all remaining intersections are obtained.
[0008] Furthermore, after obtaining the cross product, if the cross product is equal to 0, the two combined sight lines do not generate an intersection.
[0009] Furthermore, the actual intersection points are selected from all remaining intersection points by the following method: If one of the two sight lines connected by a certain intersection point has only one intersection point, then the intersection point is a real intersection point.
[0010] Furthermore, the calculating the coordinates of the actual intersection point includes: calculating the coordinates of the actual intersection point by an intersection method based on the coordinates of the two adjacent viewpoints and the direction vectors of the two sight lines associated with the actual intersection point; After calculating the coordinates of any real intersection point, delete the sight line associated with the real intersection point.
[0011] In a second aspect, the present invention provides a power grid target positioning device based on panoramic images, comprising: The target recognition module is configured to: obtain panoramic images from two adjacent viewpoints, input them into a trained target detection model, identify all targets, mark them with detection boxes, and use the bottom midpoint of the detection box corresponding to the target as the target reference point to obtain the pixel coordinates of the target reference point; The sight line calculation module is configured to: construct a sight line from the viewpoint to the target reference point based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point, calculate the direction vector of the sight line, and poll all viewpoints and target reference points to obtain all sight lines; The intersection calculation module is configured to: combine the sight lines of two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersections to obtain all remaining intersections; The target positioning module is configured to: select the actual intersection point from all remaining intersection points, calculate the coordinates of the actual intersection point and use it as the coordinates of the target to complete the power grid target positioning.
[0012] In a third aspect, the present invention provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the steps of the panoramic image-based power grid target positioning method described in any one of the first aspects are implemented.
[0013] In a fourth aspect, the present invention provides a computer system comprising: Memory, used to store computer programs / instructions; A processor is configured to execute the computer program / instructions to implement the steps of the power grid target positioning method based on panoramic images described in any one of the first aspects.
[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a power grid target positioning method, device, storage medium and system based on panoramic images. The method uses the target detection algorithm in the target detection model to extract small targets from the panoramic images, constructs the line of sight from the viewpoint to the small target, and the intersection of the two lines of sight corresponding to the two panoramic images of two adjacent viewpoints is the geographical location of the small target. In addition, the present invention automatically eliminates pseudo-intersections, obtains real intersections and calculates their coordinates, thereby realizing automatic distinction and precise positioning of multiple small target objects of the same type. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 is a flow chart of a method for power grid target positioning based on panoramic images corresponding to Example 1; Figure 2 is a flow chart of a method for power grid target positioning based on panoramic images corresponding to Example 2; Figure 3 This is a schematic diagram of a panoramic image in the process of automatically identifying a charging pile and obtaining a target reference point through a target detection model provided in Example 2; Figure 4 This is a schematic diagram of the process of calculating line of sight provided in Example 2; Figure 5 This is a schematic diagram of eliminating false intersections and screening real intersections provided in Example 2; Figure 6 This is a schematic diagram of determining target coordinates using the intersection method provided in Example 2. DETAILED DESCRIPTION
[0016] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.
[0017] Example 1
[0018] like Figure 1 As shown, this embodiment provides a method for positioning a power grid target based on panoramic images, including: After obtaining panoramic images from two adjacent viewpoints, they are input into the trained object detection model to identify all objects and mark them with detection boxes. The bottom midpoint of the detection box corresponding to the target is used as the target reference point to obtain the pixel coordinates of the target reference point. Based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point, a line of sight from the viewpoint to the target reference point is constructed, and the direction vector of the line of sight is calculated. All viewpoints and target reference points are polled to obtain all lines of sight. Combine the sight lines of two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersection points to obtain all remaining intersection points; The actual intersection points are selected from all the remaining intersection points, and the coordinates of the actual intersection points are calculated and used as the coordinates of the target to complete the power grid target positioning.
[0019] The present invention extracts small targets from panoramic images through the target detection algorithm in the target detection model, constructs the line of sight from the viewpoint to the small target, and the intersection of the two lines of sight corresponding to the two panoramic images of two adjacent viewpoints is the geographical location of the small target. In addition, the present invention automatically eliminates pseudo-intersections, obtains real intersections and calculates their coordinates, thereby realizing automatic distinction and precise positioning of multiple small target objects of the same type.
[0020] Example 2
[0021] like Figure 2 As shown, this embodiment provides a method for positioning a power grid target based on panoramic images, and its specific implementation includes the following steps: Step S1: Obtain panoramic images of two adjacent viewpoints and obtain a set of target reference points using a target detection algorithm.
[0022] Obtain Baidu Street View images of two adjacent viewpoints as panoramic images, use the trained YOLOv11 model to identify charging piles, and mark them with detection frames. The identified charging piles are as follows: Figure 3 As shown in the figure (the green box is the detection box). The bottom midpoint of the detection box is used as the target reference point to construct the target reference point set.
[0023] The viewpoint is the position from which the panoramic image is captured.
[0024] Step S2: Construct corresponding sight lines according to the panoramic images of the two viewpoints and all target reference points.
[0025] like Figure 4 As shown, the parameter information of Baidu Street View images includes: the latitude and longitude coordinates of the viewpoint, the width W and height H of the image, and the direction of the lens in the global coordinate system. The direction of the lens in the global coordinate system includes heading and pitch, where heading represents the azimuth and pitch represents the elevation. Baidu Street View is a cylindrical projection, and each pixel corresponds to a direction. To determine the direction, first calculate its direction vector in the camera coordinate system based on the principle of cylindrical projection, and then convert it into a direction vector in the global coordinate system based on the lens direction. The azimuth and elevation angles are as follows: Figure 4 As shown, according to the above method, the pixel coordinates of the target reference point are converted into a direction vector. The direction corresponding to this direction vector is used as the line of sight from the viewpoint to the target reference point. Based on this, the line of sight sets of the panoramic images of the two viewpoints are obtained. The direction vector of the line of sight is the direction vector of the target reference point in the global coordinate system. The camera coordinate system is a three-dimensional rectangular coordinate system established with the camera's optical center as the origin. The Z axis coincides with the optical axis and points in the imaging direction. The X axis is parallel to the horizontal direction of the image plane, and the Y axis is parallel to the vertical direction of the image plane. The global coordinate system is the world coordinate system in three-dimensional space.
[0026] Figure 4 In the figure, part (a) is the information display of the panoramic image, and part (b) is a schematic diagram of the global coordinate system of a certain point in part (a), where u represents the u axis of the pixel coordinate system, v represents the v axis of the pixel coordinate system, and u p Indicates the coordinate of the central pixel point (principal point) of the panoramic image on the u-axis of the pixel coordinate system, v p Indicates the coordinate of the central pixel point (phase principal point) of the panoramic image on the v axis of the pixel coordinate system, h p Indicates the camera orientation corresponding to the principal point, p T Indicates the coordinates of the top midpoint of the detection box in the pixel coordinate system, p BIndicates the coordinates of the bottom midpoint of the detection frame in the pixel coordinate system, u t Indicates the coordinate of the top midpoint of the detection box on the u-axis of the pixel coordinate system, v t Indicates the coordinate of the top midpoint of the detection box on the v-axis of the pixel coordinate system, u b Indicates the coordinate of the bottom midpoint of the detection frame on the u-axis of the pixel coordinate system, v b Indicates the coordinate of the bottom midpoint of the detection frame on the v-axis of the pixel coordinate system, x, y and z represent the three coordinate axes of the global coordinate system, θ represents the pitch angle, Ф t Indicates the azimuth of the top midpoint of the detection box, Ф b Indicates the azimuth of the bottom midpoint of the detection box.
[0027] Step S3: Combine the sight lines of the two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersections.
[0028] like Figure 5 As shown, by sight and sight Take combination as an example: traversing the sight line The direction vector in and sight The direction vector in , calculate its cross product: , determine whether the two sight lines produce a false intersection based on the m value.
[0029] If m > 0 and the target is on the right side of the travel direction, or m < 0 and the target is on the left side of the travel direction, the intersection is a pseudo-intersection. If m = 0, it means the lines are parallel and no intersection is generated. Pseudo-intersections and the associated sight lines are eliminated.
[0030] like Figure 5 As shown, two lines of sight and The cross product m>0 indicates that the intersection is a pseudo-intersection, and the intersection is removed. . Figure 5 middle, 、 、 and It’s a different perspective. and are different pseudo-intersection points, and They are two different target objects. and There are two viewpoints.
[0031] Step S4: obtaining the actual intersection point through an iterative method, and calculating the coordinates of the actual intersection point as the geographic coordinates of the small target object.
[0032] Check the intersection points that have not been eliminated to determine whether they are real intersection points: If one of the two sight lines associated with a certain intersection point is associated with only one intersection point, then the intersection point is a real intersection point. Get the first real intersection point, such as Figure 6 As shown, the geographic coordinates of the actual intersection point are calculated using the aforementioned intersection method, and the sight lines associated with the intersection point are deleted. The remaining intersection points are iteratively processed in the same way until the geographic coordinates of all the actual intersection points are obtained. The coordinates of the actual intersection points are used as the geographic coordinates of the target. Figure 5 As shown, Related sight Only associated ,therefore For real intersection points, after calculating the coordinates, delete the associated sight line and , at this time with Related pseudo-intersections It is also automatically eliminated, and then the intersection points of the remaining sights are iteratively processed to obtain the coordinates of all actual intersection points.
[0033] Figure 6 middle,( , ) is the coordinate of the first viewpoint, ( , ) is the coordinate of the second viewpoint, ( , ) is the coordinate of the target, θ b1 is the pitch angle of the first viewpoint, θ b2 is the pitch angle of the second viewpoint.
[0034] Example 3
[0035] Based on the same technical concept as Example 1, this embodiment provides a power grid target positioning device based on panoramic images, including: The target recognition module is configured to: obtain panoramic images from two adjacent viewpoints, input them into a trained target detection model, identify all targets, mark them with detection boxes, and use the bottom midpoint of the detection box corresponding to the target as the target reference point to obtain the pixel coordinates of the target reference point; The sight line calculation module is configured to: construct a sight line from the viewpoint to the target reference point based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point, calculate the direction vector of the sight line, and poll all viewpoints and target reference points to obtain all sight lines; The intersection calculation module is configured to: combine the sight lines of two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersections to obtain all remaining intersections; The target positioning module is configured to: select the actual intersection point from all remaining intersection points, calculate the coordinates of the actual intersection point and use it as the coordinates of the target to complete the power grid target positioning.
[0036] Example 4
[0037] Based on the same technical concept as Example 1, this embodiment provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, the steps of the method for power grid target positioning based on panoramic imaging provided in Example 1 are implemented as follows: After obtaining panoramic images from two adjacent viewpoints, they are input into the trained object detection model to identify all objects and mark them with detection boxes. The bottom midpoint of the detection box corresponding to the target is used as the target reference point to obtain the pixel coordinates of the target reference point. Based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point, a line of sight from the viewpoint to the target reference point is constructed, and the direction vector of the line of sight is calculated. All viewpoints and target reference points are polled to obtain all lines of sight. Combine the sight lines of two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersection points to obtain all remaining intersection points; The actual intersection points are selected from all the remaining intersection points, and the coordinates of the actual intersection points are calculated and used as the coordinates of the target to complete the power grid target positioning.
[0038] Example 5
[0039] Based on the same technical concept as Example 1, this embodiment provides a computer system, including: Memory, used to store computer programs / instructions; A processor is configured to execute the computer program / instructions to implement the steps of the power grid target positioning method based on panoramic imaging provided in Example 1: After obtaining panoramic images from two adjacent viewpoints, they are input into the trained object detection model to identify all objects and mark them with detection boxes. The bottom midpoint of the detection box corresponding to the target is used as the target reference point to obtain the pixel coordinates of the target reference point. Based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point, a line of sight from the viewpoint to the target reference point is constructed, and the direction vector of the line of sight is calculated. All viewpoints and target reference points are polled to obtain all lines of sight. Combine the sight lines of two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersection points to obtain all remaining intersection points; The actual intersection points are selected from all the remaining intersection points, and the coordinates of the actual intersection points are calculated and used as the coordinates of the target to complete the power grid target positioning.
[0040] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application 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.
[0041] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, 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 flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0042] 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.
[0043] 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 A step that specifies a function in one or more boxes.
[0044] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A power grid target positioning method based on panoramic images, characterized in that: include: After obtaining panoramic images from two adjacent viewpoints, they are input into the trained object detection model to identify all objects and mark them with detection boxes. The bottom midpoint of the detection box corresponding to the target is used as the target reference point to obtain the pixel coordinates of the target reference point. Based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point, a line of sight from the viewpoint to the target reference point is constructed, and the direction vector of the line of sight is calculated. All viewpoints and target reference points are polled to obtain all lines of sight. Combine the sight lines of two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersection points to obtain all remaining intersection points; The actual intersection points are selected from all the remaining intersection points, and the coordinates of the actual intersection points are calculated and used as the coordinates of the target to complete the power grid target positioning.
2. The method for power grid target positioning based on panoramic images according to claim 1, characterized in that: The method of constructing a sight line from the viewpoint to the target reference point based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point and calculating the direction vector of the sight line includes: Get the lens direction in the panoramic image of the viewpoint; Calculate the direction vector of the target reference point in the camera coordinate system; Converting the direction vector of the target reference point in the camera coordinate system into the direction vector of the target reference point in the global coordinate system based on the lens direction, and taking the direction corresponding to the direction vector of the target reference point in the global coordinate system as the line of sight from the viewpoint to the target reference point; The direction vector of the sight line is the direction vector of the target reference point in the global coordinate system; The camera coordinate system is a three-dimensional rectangular coordinate system established with the camera optical center as the origin. The Z axis coincides with the optical axis and points in the imaging direction. The X axis is parallel to the horizontal direction of the image plane, and the Y axis is parallel to the vertical direction of the image plane. Among them, the global coordinate system is the world coordinate system in three-dimensional space.
3. The method for power grid target positioning based on panoramic images according to claim 1, characterized in that: The sight lines of the two viewpoints are combined in pairs, and the sight line intersection analysis is performed by the vector cross product method, and the pseudo intersection points are eliminated to obtain all the remaining intersection points, including: Combine the sight lines of the two viewpoints in pairs to obtain all intersection points; Calculate the cross product of the direction vectors of the two combined sight lines; If the cross product is greater than 0 and the target is located on the right side of the travel direction, the intersection point corresponding to the cross product is a pseudo-intersection point, and the pseudo-intersection point is discarded; If the cross product is less than 0 and the target is located on the left side of the travel direction, the intersection point corresponding to the cross product is a pseudo-intersection point, and the pseudo-intersection point is discarded; After eliminating all pseudo intersections, all remaining intersections are obtained.
4. The method for power grid target positioning based on panoramic images according to claim 3, characterized in that: After obtaining the cross product, if the cross product is equal to 0, the two combined sight lines do not generate an intersection.
5. The method for power grid target positioning based on panoramic images according to claim 1, characterized in that: The actual intersection points are selected from all remaining intersection points by the following method: If one of the two sight lines associated with an intersection point is associated with only one intersection point, then the intersection point is a real intersection point.
6. The method for power grid target positioning based on panoramic images according to claim 1, characterized in that: The calculating the coordinates of the actual intersection point includes: calculating the coordinates of the actual intersection point by an intersection method based on the coordinates of the two adjacent viewpoints and the direction vectors of the two sight lines associated with the actual intersection point; After calculating the coordinates of any real intersection point, delete the sight line associated with the real intersection point.
7. A power grid target positioning device based on panoramic images, characterized in that: include: The target recognition module is configured to: obtain panoramic images from two adjacent viewpoints, input them into a trained target detection model, identify all targets, mark them with detection boxes, and use the bottom midpoint of the detection box corresponding to the target as the target reference point to obtain the pixel coordinates of the target reference point; The sight line calculation module is configured to: construct a sight line from the viewpoint to the target reference point based on the panoramic image of the viewpoint and the pixel coordinates of the target reference point, calculate the direction vector of the sight line, and poll all viewpoints and target reference points to obtain all sight lines; The intersection calculation module is configured to: combine the sight lines of two viewpoints in pairs, perform sight line intersection analysis using the vector cross product method, and eliminate false intersections to obtain all remaining intersections; The target positioning module is configured to: select the actual intersection point from all remaining intersection points, calculate the coordinates of the actual intersection point and use it as the coordinates of the target to complete the power grid target positioning.
8. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the steps of the power grid target positioning method based on panoramic images described in any one of claims 1 to 6 are implemented.
9. A computer system, characterized in that: include: Memory, used to store computer programs / instructions; A processor is used to execute the computer program / instructions to implement the steps of the power grid target positioning method based on panoramic imaging according to any one of claims 1 to 6.