Picture processing method and system based on unmanned aerial vehicle

Abstract

The invention discloses a photo processing method and system based on an unmanned aerial vehicle, and relates to the technical field of unmanned aerial vehicle processing, and the method comprises the steps: setting a photo importing rule on an operation display platform, selecting an unmanned aerial vehicle photo path, and automatically creating and naming a folder according to the photo path rule; the unmanned aerial vehicle cluster sets a route to autonomously fly according to the divided photo paths, and polls the photo paths according to a set polling task in the flight process; and positioning an archiving position according to an inspection result, and storing the picture of the unmanned aerial vehicle by using the positioned archiving position. The method provides reliable guarantee for subsequent data analysis and decision support, and is of great significance in promoting the application of the unmanned aerial vehicle technology in multiple fields such as electric power inspection and environment monitoring.

Description

Technical Field

[0001] The present invention relates to the field of drone processing technology, and in particular to a drone-based photo processing method and system. Background Art

[0002] With the rapid development of my country's economy, the society's demand for electric energy has increased dramatically, the power transmission network has continued to expand, and the mileage of transmission lines has increased significantly. The traditional power inspection method can no longer meet the current inspection needs. The initial transmission line inspection method mainly relied on manual walking along the line or using transportation, using telescopes and infrared thermal imagers and other equipment to inspect and detect line equipment and channel environment. With the continuous maturity of drone technology, drones have made significant breakthroughs in remote control, endurance time and flight quality. These technological breakthroughs have provided good conditions for the application of drones in power inspection.

[0003] At present, the existing technology is aimed at transmission lines and is not suitable for distribution lines. The transmission lines have established a three-dimensional coordinate system for the towers. The naming method is to obtain the information of the three-dimensional coordinate system when the terminal equipment takes the inspection photos to name the photos taken. This method has a great disadvantage. It is impossible to name the photos of the lines without a three-dimensional coordinate system, and the distribution lines basically have no three-dimensional coordinate system. Another disadvantage is that the invention patents that have been applied for do not involve the archiving of photos, and drone photos need to be stored in folders. Summary of the invention

[0004] In view of the problems existing in the existing drone-based photo processing methods and systems, the present invention is proposed.

[0005] Therefore, the purpose of the present invention is to provide a photo processing method and system based on drones. There are disadvantages in naming photos for routes that can no longer be processed in a three-dimensional coordinate system. The present invention adopts a photo processing method and system based on drones.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a photo processing method based on a drone, which includes setting a rule for importing photos on an operation display platform, selecting a drone photo path, and automatically creating and naming a folder according to the photo path rule; the drone cluster sets a route for autonomous flight according to the divided photo path, and inspects the photo path according to the set inspection task during the flight;

[0008] According to the inspection results, the archiving location is located, and the drone photos are stored in the located archiving location.

[0009] As a preferred solution of the drone-based photo processing method of the present invention, wherein: setting the import photo rule includes analyzing the drone operation scene according to the dynamic scene and the weak texture scene, the drone operation scene includes dynamically eliminating the positioning method, the positioning method of the weak texture scene is point-line feature fusion, and the point-line features in the image extracted by the drone binocular vision are analyzed for the error between the actual and the model, and the analysis includes estimating the drone posture by designing a positioning optimization method, establishing a drone posture optimization model, designing the feature reprojection error in the model, segmenting the dynamic content by the dynamic elimination method of Mask R-CNN, and designing the feature reprojection error in the model includes the following steps:

[0010] The feature matching relationship between the current frame and the previous frame is obtained through point and line features, and the camera motion is estimated based on the matching relationship;

[0011] When the transformation matrix from the previous frame to the current frame is not accurate enough, there is a deviation in the projection of the point and line features in the 3D space to the plane. At this time, the nonlinear least squares form of the pose optimization model is:

[0012]

[0013] Among them, ξ∈SE(3) represents the corresponding camera pose T cw The Lie algebra of , where the subscripts c and w represent the camera reference system and the world reference system, KF represents the local keyframe, P represents the local map point, and L represents the set of local map lines. represents the point feature reprojection error, represents the line feature reprojection error, The information matrix representing the reprojection error of the point, The information matrix representing the reprojection error of the line, i represents a keyframe.

[0014] As a preferred solution of the drone-based photo processing method of the present invention, wherein: automatically creating and naming a folder according to the photo path rule includes obtaining a coordinate transformation relationship according to the drone reference system and the world reference system, and the specific steps are as follows:

[0015] If i is the i-th key frame, then based on T cw ∈SE(3) UAV pose, through The 3D point of the jth feature is detected. At this time, the 3D point T cw ∈SE(3) is the UAV pose, which is transformed to the UAV coordinate system as

[0016] When passing When a 2D point on the image plane is detected, then are the two 3D endpoints of the kth line feature, which are transformed to the drone coordinate system as and are the 2D endpoints of the line features projected onto the image plane, are the 2D endpoints of the line features detected on the image plane, are the corresponding homogeneous coordinates;

[0017] The relationship between the world coordinates and pixel coordinates of the feature points and feature line endpoints in the key frame is calculated based on the coordinate transformation relationship. The specific calculation formula is:

[0018]

[0019] Among them, S j represents the jth feature point, and K is the internal parameter of the drone.

[0020] As a preferred solution of the drone-based photo processing method of the present invention, wherein: setting the route for autonomous flight includes the drone cluster selecting a flight strategy that autonomously adjusts to the terrain based on the task requirements during the inspection flight;

[0021] The flight strategy includes collecting the characteristics of the UAV operation scene, including route layout, topography and meteorological conditions, using the 3D navigation system and automatic route planning algorithm, combined with the performance parameters of the UAV cluster, to automatically generate the flight path;

[0022] According to the automatically generated flight path, the drone cluster is inspected according to the set mission requirements during the flight. The drone clusters are numbered in sequence, marked as 1, 2...n, and the time for drone cluster detection is equally divided into sub-time areas, marked as 1, 2...k; during the flight of the drone cluster's inspection of the transmission line, the flight strategy that autonomously adjusts to the terrain is selected according to the mission requirements. Multiple drone clusters share the collected information during flight and perform collaborative operations based on the shared information.

[0023] As a preferred solution of the photo processing method based on drone of the present invention, the positioning of the archive position includes eliminating the accumulated error by closed-loop detection, and the closed-loop detection includes judging whether the image feature is established at the current position according to the similarity score between the key frame and the candidate key frame in the database. If established, the image features of the current key frame and the candidate key frame are compared by dictionary matching, and the point and line similarity of the two frames is calculated. The specific calculation formula is:

[0024]

[0025] Among them, in practice, there will be a large number of zeros in the BoW feature vector, the feature vector does not include all the words in the dictionary, and similar features are filed into the same category.

[0026] As a preferred solution of the drone-based photo processing method of the present invention, the drone photos are stored, including the closed-loop detection scoring of the same category of fusion points and lines in the archive, and optimization is performed based on the closed-loop detection score. The specific scoring formula is:

[0027]

[0028] Among them, N p and N l Represents the number of point features and line features extracted from the image.

[0029] As a preferred solution of the drone-based photo processing method of the present invention, the storage includes clicking on the photo archive in the upper left corner of the homepage after logging in, selecting the folder where the photos taken by the drone are located, selecting the target folder, selecting the folder where the photos will be stored, and executing the archiving operation. Clicking cancel will not execute the archiving operation, otherwise, the archiving operation is executed.

[0030] In the second aspect, an embodiment of the present invention provides a photo processing system based on a drone, which includes: an import module, which sets import photo rules on an operation display platform, selects a drone photo path, and automatically creates and names folders according to the photo path rules; an inspection module, in which a drone cluster sets a route according to the divided photo path and flies autonomously, and inspects the photo path according to the set inspection tasks during the flight; a storage module, which locates the archiving location according to the inspection results, and stores the drone photos in the located archiving location.

[0031] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, any step of the above-mentioned drone-based photo processing method is implemented.

[0032] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned drone-based photo processing method is implemented.

[0033] The beneficial effects of the present invention are as follows: the present invention provides a photo processing method and system based on drones, which effectively improves the automation level and data management efficiency of drone inspection work through intelligently setting photo import rules, autonomous flight inspection of drone clusters, and efficient photo storage processes. In particular, in drone operations under dynamic scenes and weak texture scenes, the accuracy of drone pose estimation is significantly improved by introducing a positioning method that combines point and line features and a dynamic content segmentation technology based on Mask R-CNN. In addition, the method also realizes collaborative operations between drone clusters, and can autonomously adjust flight strategies according to terrain and meteorological conditions, further enhancing the flexibility and adaptability of the drone inspection system. Finally, through closed-loop detection and archiving location optimization, the accuracy and timeliness of drone-collected data are ensured, providing reliable guarantees for subsequent data analysis and decision support. The invention is of great significance for promoting the application of drone technology in multiple fields such as power inspection and environmental monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0035] Figure 1 A specific flow chart of a drone-based photo processing method and system provided for one embodiment of the present invention.

[0036] Figure 2 A system directory diagram of a drone-based photo processing method and system provided by an embodiment of the present invention.

[0037] Figure 3 A schematic diagram of system login for a drone-based photo processing method and system provided by an embodiment of the present invention.

[0038] Figure 4 A schematic diagram of photo archiving of a drone-based photo processing method and system provided by an embodiment of the present invention.

[0039] Figure 5 A schematic diagram of selecting a folder in a drone-based photo processing method and system provided by an embodiment of the present invention.

[0040] Figure 6 A schematic diagram of a patrol of a drone-based photo processing method and system provided by an embodiment of the present invention.

[0041] Figure 7A newly added schematic diagram of the configuration of a drone-based photo processing method and system provided for one embodiment of the present invention.

[0042] Figure 8 A schematic diagram of deletion settings of a drone-based photo processing method and system provided by an embodiment of the present invention.

[0043] Fig. 9 The present invention provides an operation diagram of a drone-based photo processing method and system according to an embodiment of the present invention.

[0044] Fig.10 A schematic diagram of successful archiving of a drone-based photo processing method and system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0046] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0047] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0048] The present invention is described in detail with reference to the schematic diagram. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0049] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0050] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0051] Example 1

[0052] Reference Figures 1 to 10 , which is the first embodiment of the present invention, provides a photo processing method based on a drone, comprising:

[0053] S1: Set the import photo rules on the operation display platform, select the drone photo path, and automatically create and name the folder according to the photo path rules.

[0054] Among them, setting the rules for importing photos includes analyzing the drone operation scene according to dynamic scenes and weak texture scenes. The drone operation scene includes dynamically eliminating the positioning method. The positioning method of the weak texture scene is point-line feature fusion. The point-line features in the image extracted by the drone binocular vision are analyzed for the error between the actual and the model. The analysis includes estimating the drone posture by designing a positioning optimization method, establishing a drone posture optimization model, designing the feature reprojection error in the model, segmenting the dynamic content by the dynamic elimination method of Mask R-CNN, and designing the feature reprojection error in the model. The following steps are included:

[0055] The feature matching relationship between the current frame and the previous frame is obtained through point and line features, and the camera motion is estimated based on the matching relationship;

[0056] When the transformation matrix from the previous frame to the current frame is not accurate enough, there is a deviation in the projection of the point and line features in the 3D space to the plane. At this time, the nonlinear least squares form of the pose optimization model is:

[0057]

[0058] Among them, ξ∈SE(3) represents the corresponding camera pose Tcw The Lie algebra of , where the subscripts c and w represent the camera reference system and the world reference system, KF represents the local keyframe, P represents the local map point, and L represents the set of local map lines. represents the point feature reprojection error, represents the line feature reprojection error, The information matrix representing the reprojection error of the point, The information matrix representing the reprojection error of the line, i represents a keyframe.

[0059] S1.1: Automatically create and name folders according to the photo path rules, including obtaining the coordinate transformation relationship according to the drone reference system and the world reference system. The specific steps are as follows:

[0060] If i is the i-th key frame, then based on T cw ∈SE(3) UAV pose, through The 3D point of the jth feature is detected. At this time, the 3D point T cw ∈SE(3) is the UAV pose, which is transformed to the UAV coordinate system as

[0061] When passing When a 2D point on the image plane is detected, then are the two 3D endpoints of the kth line feature, which are transformed to the drone coordinate system as and are the 2D endpoints of the line features projected onto the image plane, are the 2D endpoints of the line features detected on the image plane, are the corresponding homogeneous coordinates;

[0062] The relationship between the world coordinates and pixel coordinates of the feature points and feature line endpoints in the key frame is calculated based on the coordinate transformation relationship. The specific calculation formula is:

[0063]

[0064] Among them, S j represents the jth feature point, and K is the internal parameter of the drone.

[0065] S2: The drone cluster sets the route and flies autonomously according to the divided photo paths, and inspects the photo paths according to the set inspection tasks during the flight.

[0066] Among them, the route setting autonomous flight includes the process of the drone swarm selecting the flight strategy that autonomously adjusts to the terrain according to the mission requirements during the inspection flight.

[0067] The flight strategy includes collecting the characteristics of the UAV operation scene, including route layout, topography and meteorological conditions, using the 3D navigation system and automatic route planning algorithm, combined with the performance parameters of the UAV cluster, to automatically generate the flight path;

[0068] According to the automatically generated flight path, the drone cluster is inspected according to the set mission requirements during the flight. The drone clusters are numbered in sequence, marked as 1, 2...n, and the time for drone cluster detection is equally divided into sub-time areas, marked as 1, 2...k; during the flight of the drone cluster's inspection of the transmission line, the flight strategy that autonomously adjusts to the terrain is selected according to the mission requirements. Multiple drone clusters share the collected information during flight and perform collaborative operations based on the shared information.

[0069] S3: According to the inspection results, locate the archive location, and use the located archive location to store the drone photos.

[0070] Among them, locating the archive position includes using closed-loop detection to eliminate the cumulative error. The closed-loop detection includes judging whether the image feature is established at the current position according to the similarity score between the key frame and the candidate key frame in the database. If established, the image features of the current key frame and the candidate key frame are compared through dictionary matching to calculate the point and line similarity of the two frames. The specific calculation formula is:

[0071]

[0072] Among them, in practice, there will be a large number of zeros in the BoW feature vector, the feature vector does not include all the words in the dictionary, and similar features are filed into the same category.

[0073] S3.1: Storing drone photos includes merging points and lines of the same category in the archive for closed-loop detection scoring, and optimizing based on the closed-loop detection scoring. The specific scoring formula is:

[0074]

[0075] Among them, N p and N l Represents the number of point features and line features extracted from the image.

[0076] Preferably, storage includes clicking on photo archiving in the upper left corner of the homepage after logging in, selecting the folder where the photos taken by the drone are located, selecting the target folder, selecting the folder where the photos will be stored, executing the archiving operation, clicking Cancel to not execute the archiving operation, otherwise, executing the archiving operation.

[0077] In a preferred embodiment, a drone-based photo processing method and system includes an import module, which sets import photo rules on an operation display platform, selects drone photo paths, and automatically creates and names folders according to the photo path rules; an inspection module, in which a drone cluster sets routes according to the divided photo paths and flies autonomously, and inspects the photo paths according to the set inspection tasks during the flight; and a storage module, which locates the archiving location according to the inspection results, and uses the located archiving location to store the drone photos.

[0078] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0079] In summary, the present invention provides a photo processing method and system based on drones, which effectively improves the automation level and data management efficiency of drone inspection work through intelligent setting of photo import rules, autonomous flight inspection of drone clusters, and efficient photo storage process. In particular, in drone operations under dynamic scenes and weak texture scenes, the accuracy of drone pose estimation is significantly improved by introducing a positioning method of point-line feature fusion and dynamic content segmentation technology based on Mask R-CNN. In addition, the method also realizes collaborative operations between drone clusters, and can autonomously adjust flight strategies according to terrain and meteorological conditions, further enhancing the flexibility and adaptability of the drone inspection system. Finally, through closed-loop detection and archiving position optimization, the accuracy and timeliness of drone collected data are ensured, providing reliable guarantees for subsequent data analysis and decision support. This invention is of great significance for promoting the application of drone technology in multiple fields such as power inspection and environmental monitoring.

[0080] Example 2

[0081] Reference Figures 1 to 10, which is the second embodiment of the present invention, and this embodiment provides a photo processing method based on a drone. In order to verify the beneficial effects of the present invention, a scientific demonstration is carried out through simulation experiments.

[0082] In the implementation experiment of the present invention, a transmission line area with an area of ​​about 10 square kilometers was selected as the experimental site. The area contains complex terrain and various vegetation covers. At the same time, multiple dynamic scenes (such as moving vehicles) and weak texture scenes (such as smooth walls) were set. A cluster of 5 drones was used in the experiment. Each drone was equipped with a binocular vision sensor and installed with the photo processing system proposed by the present invention. At the beginning of the experiment, the import photo rules were set by operating the display platform, and 5 folders were automatically created and named for storing photos taken by different drones. The drone cluster flew autonomously according to the preset route. During the flight, the designated path was inspected in detail according to the set inspection tasks. The total flight time was about 2 hours, and a total of about 1,500 high-definition pictures were taken. After the inspection was completed, the system successfully eliminated the cumulative error using closed-loop detection technology, and accurately filed all photos taken by the drone into the corresponding folder, with an archiving accuracy of more than 98%. By analyzing the reprojection error of point and line features, the drone posture model was optimized, and the posture estimation error was reduced from the initial average of 3.5 meters to less than 0.5 meters, greatly improving the accuracy of the drone inspection operation. The experimental results show that the present invention can effectively improve the efficiency and data processing capability of UAV inspection work and has high practical value. The experimental data are shown in Table 1 below:

[0083] Table 1 Experimental data table

[0084] Experimental Project Specific content Experimental site area About 10 square kilometers Topographic features Contains complex terrain and multiple vegetation covers Dynamic Scenes Such as moving vehicles Weak texture scene Such as smooth wall Number of drones 5 Each drone configuration Binocular vision sensor Folder creation 5 folders are automatically created and named Total flight time About 2 hours Total number of pictures taken About 1500 Filing accuracy More than 98% Initial pose estimation error Average 3.5 meters Optimized pose estimation error Within 0.5 meters

[0085] Table 1 clearly shows the basic information and key data of the experiment, which is helpful to intuitively understand the practical application effect of the present invention. The comparison between the present invention and the prior art is shown in Table 2 below:

[0086] Table 2 Comparison between the present invention and the prior art

[0087]

[0088] It can be seen from Table 2 that the present invention has significant advantages in many aspects, especially in improving the flexibility of UAV inspection, accuracy and efficiency of data processing.

[0089] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A photo processing method based on drone, characterized by: include, Set up the rules for importing photos on the operation display platform, select the drone photo path, and automatically create and name folders based on the photo path rules; The drone swarm sets the route and flies autonomously according to the divided photo paths, and inspects the photo paths according to the set inspection tasks during the flight; According to the inspection results, the archiving location is located, and the drone photos are stored in the located archiving location.

2. The drone-based photo processing method according to claim 1, characterized in that: Setting the rules for importing photos includes analyzing the UAV operation scene according to dynamic scenes and weak texture scenes. The UAV operation scene includes dynamic elimination of the positioning method. The positioning method of the weak texture scene is point-line feature fusion. The point-line features in the image extracted by the UAV binocular vision are analyzed for the error between the actual and the model. The analysis includes estimating the UAV posture by designing a positioning optimization method, establishing a UAV posture optimization model, designing the feature reprojection error in the model, and segmenting the dynamic content by the dynamic elimination method of Mask R-CNN. The feature reprojection error in the design model includes the following steps: The feature matching relationship between the current frame and the previous frame is obtained through point and line features, and the camera motion is estimated based on the matching relationship; When the transformation matrix from the previous frame to the current frame is not accurate enough, there is a deviation in the projection of the point and line features in the 3D space to the plane. At this time, the nonlinear least squares form of the pose optimization model is: Among them, ξ∈SE(3) represents the corresponding camera pose T cw The Lie algebra of , where the subscripts c and w represent the camera reference system and the world reference system, KF represents the local keyframe, P represents the local map point, and L represents the set of local map lines. represents the point feature reprojection error, represents the line feature reprojection error, The information matrix representing the reprojection error of the point, The information matrix representing the reprojection error of the line, i represents a keyframe.

3. The drone-based photo processing method according to claim 2, characterized in that: Automatically create and name folders based on photo path rules, including obtaining coordinate transformation relationships based on the drone reference system and the world reference system. The specific steps are as follows: If i is the i-th key frame, then based on T cw ∈SE(3) UAV pose, through The 3D point of the jth feature is detected. At this time, the 3D point T cw ∈SE(3) is the UAV pose, which is transformed to the UAV coordinate system as When passing When a 2D point on the image plane is detected, then are the two 3D endpoints of the kth line feature, which are transformed to the drone coordinate system as and are the 2D endpoints of the line features projected onto the image plane, are the 2D endpoints of the line features detected on the image plane, are the corresponding homogeneous coordinates; The relationship between the world coordinates and pixel coordinates of the feature points and feature line endpoints in the key frame is calculated based on the coordinate transformation relationship. The specific calculation formula is: Among them, S j represents the jth feature point, and K is the internal parameter of the drone.

4. The drone-based photo processing method according to claim 3, characterized in that: Setting route autonomous flight includes the process of UAV swarm based on patrol flight, in which the UAV swarm selects a flight strategy that autonomously adjusts to the terrain according to mission requirements; The flight strategy includes collecting the characteristics of the UAV operation scene, including route layout, topography and meteorological conditions, using the 3D navigation system and automatic route planning algorithm, combined with the performance parameters of the UAV cluster, to automatically generate the flight path; According to the automatically generated flight path, the drone cluster is inspected according to the set mission requirements during the flight. The drone clusters are numbered in sequence, marked as 1, 2...n, and the time for drone cluster detection is equally divided into sub-time areas, marked as 1, 2...k; during the flight of the drone cluster's inspection of the transmission line, the flight strategy that autonomously adjusts to the terrain is selected according to the mission requirements. Multiple drone clusters share the collected information during flight and perform collaborative operations based on the shared information.

5. The drone-based photo processing method according to claim 4, characterized in that: The location of the archive includes the use of closed-loop detection to eliminate the accumulated error. The closed-loop detection includes judging whether the image feature is established at the current position according to the similarity score between the key frame and the candidate key frame in the database. If established, the image features of the current key frame and the candidate key frame are compared through dictionary matching to calculate the point and line similarity of the two frames. The specific calculation formula is: Among them, in practice, there will be a large number of zeros in the BoW feature vector, the feature vector does not include all the words in the dictionary, and similar features are filed into the same category.

6. The drone-based photo processing method according to claim 5, characterized in that: Storing drone photos includes merging points and lines of the same category in the archive for closed-loop detection scoring, and optimizing based on the closed-loop detection scoring. The specific scoring formula is: Among them, N p and N l Represents the number of point features and line features extracted from the image.

7. The drone-based photo processing method according to claim 6, characterized in that: Storage includes logging in and clicking on photo archiving in the upper left corner of the homepage to enter, selecting the folder where the photos taken by the drone are located, selecting the target folder, selecting the folder where the photos will be stored, and executing the archiving operation. Clicking Cancel will not execute the archiving operation, otherwise, the archiving operation will be executed.

8. A drone-based photo processing system, based on the drone-based photo processing method according to any one of claims 1 to 7, characterized in that: include, The import module sets the rules for importing photos on the operation display platform, selects the drone photo path, and automatically creates and names folders according to the photo path rules; Inspection module: its drone cluster sets the route and flies autonomously according to the divided photo paths, and inspects the photo paths according to the set inspection tasks during the flight; The storage module locates the archiving location according to the inspection results, and uses the located archiving location to store the photos taken by the drone.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of any one of claims 1 to 7 of the drone-based photo processing method are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of any one of claims 1 to 7 of the drone-based photo processing method are implemented.

Images