Power monitoring method and system based on unmanned aerial vehicle cruise

Abstract

The invention relates to a power monitoring method and system based on unmanned aerial vehicle cruise, and the method comprises the following steps: enabling an unmanned aerial vehicle to cruise according to an initial cruise path, so as to shoot a power transmission line, and obtaining an initial power transmission line picture; acquiring a cruise path deviation value of the unmanned aerial vehicle by identifying the position of the power transmission line on the initial power transmission line picture; the cruise path of the unmanned aerial vehicle is corrected according to the cruise path deviation value so that the unmanned aerial vehicle can cruise along the extension track of the power transmission line; the unmanned aerial vehicle with the corrected cruise path shoots the power transmission line to obtain a corrected power transmission line picture; constructing a power transmission line foreign matter identification model based on a computer vision identification method; identifying defects of the power transmission line through a power transmission line foreign matter identification model to obtain an identification result; judging whether the power transmission line is abnormal according to the identification result; according to the invention, the defect inspection of the power transmission line is efficiently and accurately realized, and the problems of large workload and relatively low efficiency of manual inspection of the power transmission line are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring, and in particular to a power monitoring method and system based on unmanned aerial vehicle cruising. Background Art

[0002] The distribution system is an important part of the power system, and the prevention and maintenance of distribution system network failures are very important. In order to detect defects in the distribution system network as early as possible and effectively plan the required maintenance activities, it is necessary to regularly inspect the distribution network.

[0003] Since the transmission lines in the distribution network are widely distributed, the traditional manual inspection method has great labor intensity, but the inspection efficiency is low. Helicopter inspection is also one of the existing methods for inspecting and maintaining the transmission lines of the distribution network. The helicopter inspection method has high detection accuracy, but the maintenance cost is high and there are safety issues. In view of the problem of heavy workload and low efficiency of the above-mentioned transmission line defect inspection method, no effective solution has been proposed so far. Summary of the invention

[0004] In order to solve the technical problems in the prior art of transmission line defect inspection methods such as large workload and low efficiency, the present invention provides a power monitoring method and system based on drone cruising.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] A power monitoring method based on drone cruising includes the following steps:

[0007] S1. Creating an initial cruising path based on the location information of the power transmission line;

[0008] S2, allowing the drone to cruise along the initial cruise path to photograph the power transmission line to obtain an initial transmission line photo;

[0009] S3. Obtaining a cruising path deviation value of the UAV by identifying the position of the power transmission line on the initial transmission line photo;

[0010] S4. Correcting the cruise path of the UAV according to the cruise path deviation value, so that the UAV cruises along the extended trajectory of the power transmission line;

[0011] S5, allowing the drone after the corrected cruising path to photograph the power transmission line to obtain a corrected transmission line photo; wherein the power transmission line has one or more, and when there is only one power transmission line, the camera only collects the corrected transmission line photo of the one power transmission line;

[0012] When there are multiple power transmission lines, the cameras are provided in multiple numbers, and the multiple cameras correspond to the multiple power transmission lines one by one. The initial transmission line photo is taken by one of the cameras, and the revised transmission line photo includes photos taken by multiple cameras; S6, constructing a transmission line foreign body recognition model based on a computer vision recognition method;

[0013] S7, identifying defects of the power transmission line by inputting the corrected transmission line photo into the transmission line foreign body recognition model to obtain a recognition result;

[0014] S8. Determine whether the power transmission line is abnormal based on the identification result.

[0015] The beneficial effects of the present invention are: using computer vision recognition to calculate the cruise path deviation value, and correcting the cruise path according to the cruise path deviation value, and using computer vision recognition to identify defects of the transmission line after correcting the cruise path, such as foreign matter attachment defects, transmission line fatigue damage and other defects; improving the efficient and accurate defect inspection of the transmission line, and avoiding the problem of large workload and low efficiency of manual transmission line inspection.

[0016] Based on the above technical solution, the present invention can also be improved as follows.

[0017] Furthermore, the initial UAV cruising path is created based on the location information of the power transmission line, which specifically includes the following steps:

[0018] S101, obtaining the position coordinates and altitudes of both ends of the power transmission line;

[0019] S102, creating an initial flight instruction according to the position coordinates and altitude of both ends of the power transmission line;

[0020] S103: construct the initial cruise path using the initial flight instruction.

[0021] Further, the drone is allowed to cruise along the initial cruise path to photograph the power transmission line to obtain an initial transmission line photo, which specifically includes the following steps:

[0022] S201, allowing the UAV to cruise along an initial cruise path, so that the UAV is located above the power transmission line for cruising;

[0023] S202, allowing the drone to take photos vertically downward while cruising above the power transmission line to obtain the initial transmission line photo.

[0024] Further, by identifying the position of the power transmission line on the initial transmission line photo, obtaining the cruising path deviation value of the UAV specifically includes the following steps:

[0025] S301, establishing a three-dimensional rectangular coordinate system with the focus center of the camera as the origin, the optical axis of the camera as the Z axis, and the motion direction of the drone as the X axis, to obtain a camera coordinate system;

[0026] S302, establishing a three-dimensional rectangular coordinate system with the intersection of the optical axis of the camera and the image plane of the initial transmission line photo as the coordinate origin, the optical axis of the camera as the Z axis, and the projection of the X axis of the camera coordinate system on the initial transmission line photo as the X axis to obtain an image coordinate system;

[0027] S303, calculating the Y coordinate value of the center point of the power transmission line in the initial transmission line photo passing through the Y axis of the image coordinate system in the image coordinate system;

[0028] S304, using the external parameter matrix of the camera to convert the Y coordinate value of the center point of the power transmission line in the image coordinate system to the camera coordinate system, to obtain the cruise path deviation value.

[0029] Further, calculating the Y coordinate value of the center point of the power transmission line in the initial transmission line photo passing through the Y axis of the image coordinate system in the image coordinate system specifically includes the following steps:

[0030] The initial transmission line photo is subjected to grayscale processing to obtain an initial transmission line grayscale image;

[0031] Acquire Y coordinate values ​​of a plurality of pixel points on the Y axis of the image coordinate system in the initial transmission line grayscale image to obtain a plurality of pixel Y coordinate values;

[0032] Binarize each pixel point on the Y axis of the image coordinate system to obtain a plurality of corresponding pixel point grayscale values;

[0033] Calculate the difference between two pixel points whose grayscale values ​​are within the preset range in the Y coordinate direction to obtain the central pixel difference;

[0034] The sum of the central pixel difference and the Y coordinate value of the near origin is calculated to obtain the Y coordinate value of the center point of the power transmission line in the image coordinate system; wherein the near origin is a pixel point close to the origin of the image coordinate system among the two pixel points whose grayscale values ​​are within a preset range.

[0035] Further, the cruise path of the UAV is corrected according to the cruise path deviation value, specifically comprising the following steps:

[0036] Using the GPS navigation system or the Beidou navigation system to construct the position coordinate system of the UAV;

[0037] The cruise path deviation value is converted into the position coordinate system using the camera conversion matrix of the drone to obtain a navigation deviation value;

[0038] The cruising path of the UAV is corrected according to the navigation deviation value.

[0039] Furthermore, based on the computer vision recognition method, a transmission line foreign body recognition model is constructed, which specifically includes the following steps:

[0040] S601. Constructing an initial visual recognition model based on computer visual recognition methods;

[0041] S602, collecting a plurality of abnormal images of the power transmission line to obtain a plurality of initial training images;

[0042] S603, preprocessing the multiple initial training images to remove noise in each of the initial training images to obtain multiple formal training images;

[0043] S604: Use the plurality of formal training images to train the initial visual recognition model to obtain the transmission line foreign object recognition model.

[0044] Further, the defects of the power transmission line are identified by inputting the corrected transmission line photo into the transmission line foreign body identification model to obtain an identification result, which specifically includes the following steps:

[0045] S701, inputting the modified transmission line photo into the transmission line foreign body recognition model, so as to recognize feature vectors in the modified transmission line photo through the transmission line foreign body recognition model, and obtain multiple feature vectors;

[0046] S702. Calculate the feature matching degree between each of the feature vectors and the plurality of abnormal feature vectors respectively using an approximate nearest neighbor matching method to obtain a plurality of feature matching degree values; wherein the recognition result includes the plurality of feature matching degree values.

[0047] Further, judging whether the power transmission line is abnormal according to the recognition result, the specific steps are as follows:

[0048] Each of the feature matching values ​​is compared with a preset threshold. If any of the feature matching values ​​is greater than or equal to the preset threshold, the power transmission line is judged to be abnormal; if all of the feature matching values ​​are less than the preset threshold, the power transmission line is judged to be normal.

[0049] In order to solve the above technical problems, the present invention also provides a power monitoring system based on drone cruising, and its specific technical contents are as follows:

[0050] A power monitoring system based on drone cruising, using the above-mentioned power monitoring method based on drone cruising to monitor power transmission lines, comprises:

[0051] A controller, configured to create an initial cruise path according to the location information of the power transmission line, and to create an initial control instruction according to the initial cruise path;

[0052] A drone is connected to the controller through wireless communication, and is used to cruise the power transmission line under the control of the initial control instruction to photograph the power transmission line to obtain an initial transmission line photo;

[0053] A data processing server establishes a communication connection with the controller, and is used to obtain the initial power transmission line photo through the controller; obtain the cruise path deviation value of the UAV by identifying the position of the power transmission line on the initial power transmission line photo; correct the cruise path of the UAV according to the cruise path deviation value, and transmit the corrected cruise path of the UAV to the controller;

[0054] The controller is also used to generate a correction control instruction according to the corrected cruising path of the UAV;

[0055] The drone is also used to cruise along the extension trajectory of the power transmission line under the control of the revised control instruction to photograph the power transmission line and obtain a revised transmission line photo;

[0056] The data processing server is also used to obtain the corrected transmission line photo through the controller; construct a transmission line foreign object recognition model based on a computer vision recognition method; identify the defects of the power transmission line by inputting the corrected transmission line photo into the transmission line foreign object recognition model to obtain a recognition result; and determine whether the power transmission line is abnormal based on the recognition result. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a flow chart of a power monitoring method based on drone cruising in an embodiment of the present invention;

[0058] Figure 2 The figure is a schematic diagram of the structure of a power monitoring system based on drone cruising in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0060] like Figure 1 As shown, this embodiment provides a power monitoring method based on drone cruising, comprising the following steps:

[0061] S1. Creating an initial cruising path based on the location information of the power transmission line;

[0062] The initial UAV cruising path is created based on the location information of the power transmission lines, which includes the following steps:

[0063] S101, obtaining the position coordinates and altitudes of both ends of the power transmission line;

[0064] S102, creating an initial flight instruction according to the position coordinates and altitude of both ends of the power transmission line;

[0065] S103, constructing the initial cruise path using the initial flight instruction. Specifically, the initial cruise path is mainly to let the drone fly according to a predetermined flight trajectory and altitude, so as to use the camera on the drone to shoot the power transmission line for video or image.

[0066] S2, allowing the drone to cruise along the initial cruise path to photograph the power transmission line to obtain an initial transmission line photo;

[0067] Allowing the drone to cruise along the initial cruise path to photograph the power transmission line to obtain an initial transmission line photo specifically includes the following steps:

[0068] S201, allowing the UAV to cruise along the initial cruising path so that the UAV is located above the power transmission line for cruising; specifically, the power transmission line is a single transmission line, and one UAV photographs one power transmission line or one UAV carries multiple cameras, and the multiple cameras photograph multiple power transmission lines respectively. For example, when photographing a three-phase high-voltage transmission line, three UAVs can be used to photograph the three high-voltage transmission lines respectively; when a multi-camera solution is adopted, three cameras are set to correspond to the three high-voltage transmission lines respectively, and the three cameras can be moved on the UAV individually to change the position of each camera on the UAV.

[0069] S202, allowing the drone to take photos vertically downward while cruising above the power transmission line to obtain the initial transmission line photo.

[0070] S3. Obtaining a cruising path deviation value of the UAV by identifying the position of the power transmission line on the initial transmission line photo;

[0071] By identifying the position of the power transmission line on the initial transmission line photo, obtaining the cruising path deviation value of the UAV specifically includes the following steps:

[0072] S301, establishing a three-dimensional rectangular coordinate system with the focus center of the camera as the origin, the optical axis of the camera as the Z axis, and the motion direction of the drone as the X axis, to obtain a camera coordinate system;

[0073] S302, establishing a three-dimensional rectangular coordinate system with the intersection of the optical axis of the camera and the image plane of the initial transmission line photo as the coordinate origin, the optical axis of the camera as the Z axis, and the projection of the X axis of the camera coordinate system on the initial transmission line photo as the X axis to obtain an image coordinate system;

[0074] S303, calculating the Y coordinate value of the center point of the power transmission line in the initial transmission line photo passing through the Y axis of the image coordinate system in the image coordinate system;

[0075] S304, using the external parameter matrix of the camera to convert the Y coordinate value of the center point of the power transmission line in the image coordinate system to the camera coordinate system, to obtain the cruise path deviation value.

[0076] Calculating the Y coordinate value of the center point of the power transmission line in the initial transmission line photo passing through the Y axis of the image coordinate system in the image coordinate system specifically includes the following steps:

[0077] The initial transmission line photo is subjected to grayscale processing to obtain an initial transmission line grayscale image;

[0078] Acquire Y coordinate values ​​of a plurality of pixel points on the Y axis of the image coordinate system in the initial transmission line grayscale image to obtain a plurality of pixel Y coordinate values;

[0079] Binarize each pixel point on the Y axis of the image coordinate system to obtain a plurality of corresponding pixel point grayscale values;

[0080] Calculate the difference between two pixel points whose grayscale values ​​are within the preset range in the Y coordinate direction to obtain the central pixel difference;

[0081] The sum of the central pixel difference and the Y coordinate value of the near origin is calculated to obtain the Y coordinate value of the center point of the power transmission line in the image coordinate system; wherein the near origin is a pixel point close to the origin of the image coordinate system among the two pixel points whose grayscale values ​​are within a preset range.

[0082] S4. Correcting the cruise path of the UAV according to the cruise path deviation value, so that the UAV cruises along the extended trajectory of the power transmission line;

[0083] Correcting the cruise path of the UAV according to the cruise path deviation value specifically includes the following steps:

[0084] S401, constructing a position coordinate system of the UAV using a GPS navigation system or a Beidou navigation system;

[0085] S402, using the camera conversion matrix of the UAV to convert the cruise path deviation value into the position coordinate system to obtain a navigation deviation value;

[0086] S403, correcting the cruise path of the drone according to the navigation deviation value. When multiple cameras are set on the drone, after the cruise path of the drone is corrected by the navigation deviation value, each camera can also change its position according to the Y coordinate value of the center point of the power transmission line corresponding to itself in the image coordinate system to adapt to the corrected cruise path of the drone. It is convenient to correct the cruise path according to the initial power transmission line photo of one camera, and the remaining cameras adjust the shooting position by adjusting their own positions on the drone.

[0087] S5, allowing the drone after the corrected cruising path to photograph the power transmission line to obtain a corrected transmission line photo; wherein the power transmission line has one or more, and when there is only one power transmission line, the camera only collects the corrected transmission line photo of the one power transmission line;

[0088] When there are multiple power transmission lines, multiple cameras are provided, and the multiple cameras correspond one-to-one to the multiple power transmission lines. The initial transmission line photo is taken by one of the cameras, and the revised transmission line photo includes photos taken by multiple cameras; except for the camera that takes the photo of the initial transmission line, the remaining cameras change their shooting positions by moving their relative positions with the drone to adapt to the revised cruise path.

[0089] S6. Based on computer vision recognition method, a foreign body recognition model for transmission line is constructed;

[0090] Based on computer vision recognition method, a foreign body recognition model for transmission line is constructed, which includes the following steps:

[0091] S601. Constructing an initial visual recognition model based on computer visual recognition methods;

[0092] S602, collecting a plurality of abnormal images of the power transmission line to obtain a plurality of initial training images;

[0093] S603, preprocessing the multiple initial training images to remove noise in each of the initial training images to obtain multiple formal training images;

[0094] S604: Use the plurality of formal training images to train the initial visual recognition model to obtain the transmission line foreign object recognition model.

[0095] S7, identifying defects of the power transmission line by inputting the corrected transmission line photo into the transmission line foreign body recognition model to obtain a recognition result;

[0096] The defects of the power transmission line are identified by inputting the corrected transmission line photo into the transmission line foreign body identification model to obtain an identification result, which specifically includes the following steps:

[0097] S701, inputting the modified transmission line photo into the transmission line foreign body recognition model, so as to recognize feature vectors in the modified transmission line photo through the transmission line foreign body recognition model, and obtain multiple feature vectors;

[0098] S702. Calculate the feature matching degree between each of the feature vectors and the plurality of abnormal feature vectors respectively using an approximate nearest neighbor matching method to obtain a plurality of feature matching degree values; wherein the recognition result includes the plurality of feature matching degree values.

[0099] S8. Determine whether the power transmission line is abnormal based on the identification result.

[0100] Judging whether the power transmission line is abnormal according to the recognition result, the specific steps are:

[0101] Each of the feature matching values ​​is compared with a preset threshold. If any of the feature matching values ​​is greater than or equal to the preset threshold, the power transmission line is judged to be abnormal; if all of the feature matching values ​​are less than the preset threshold, the power transmission line is judged to be normal.

[0102] The embodiment of the present invention uses computer vision recognition to calculate the cruise path deviation value, and corrects the cruise path according to the cruise path deviation value, and uses computer vision recognition to identify defects of the transmission line after correcting the cruise path, such as foreign matter attachment defects, transmission line fatigue damage and other defects; thereby improving the efficient and accurate defect inspection of the transmission line and avoiding the problem of heavy workload and low efficiency of manual transmission line inspection.

[0103] like Figure 2 As shown, the drone takes photos above the power transmission line, which is a transmission line between two adjacent power towers, that is, only the two ends of the power transmission line are provided with fixed objects, and there are no fixed objects or power towers on the power transmission line. In other embodiments, a power monitoring system based on drone cruising is also provided, and its specific technical contents are as follows:

[0104] A power monitoring system based on drone cruising, using the above-mentioned power monitoring method based on drone cruising to monitor power transmission lines, comprises:

[0105] A controller, configured to create an initial cruise path according to the location information of the power transmission line, and to create an initial control instruction according to the initial cruise path;

[0106] A drone is connected to the controller through wireless communication, and is used to cruise the power transmission line under the control of the initial control instruction to photograph the power transmission line to obtain an initial transmission line photo;

[0107] A data processing server establishes a communication connection with the controller, and is used to obtain the initial power transmission line photo through the controller; obtain the cruise path deviation value of the UAV by identifying the position of the power transmission line on the initial power transmission line photo; correct the cruise path of the UAV according to the cruise path deviation value, and transmit the corrected cruise path of the UAV to the controller;

[0108] The controller is also used to generate a correction control instruction according to the corrected cruising path of the UAV;

[0109] The drone is also used to cruise along the extension trajectory of the power transmission line under the control of the revised control instruction to photograph the power transmission line and obtain a revised transmission line photo;

[0110] The data processing server is also used to obtain the corrected transmission line photo through the controller; construct a transmission line foreign object recognition model based on a computer vision recognition method; identify the defects of the power transmission line by inputting the corrected transmission line photo into the transmission line foreign object recognition model to obtain a recognition result; and determine whether the power transmission line is abnormal based on the recognition result.

[0111] The cruise of the drone can be controlled by the controller, or the controller can send a control program or instruction to the drone, so that the drone can cruise according to the control program or instruction. The data processing server can be a cloud server, and the controller establishes a communication connection with the data processing server through a mutual benefit network. The controller and the data processing server can also be connected to each other through a local area network, and the controller and the drone are connected by wireless communication.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the concept and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A power monitoring method based on drone cruising, characterized in that: The steps include: S1. Creating an initial cruising path based on the location information of the power transmission line; S2, allowing the drone to cruise along the initial cruise path to photograph the power transmission line to obtain an initial transmission line photo; S3. Obtaining a cruising path deviation value of the UAV by identifying the position of the power transmission line on the initial transmission line photo; S4. Correcting the cruise path of the UAV according to the cruise path deviation value, so that the UAV cruises along the extended trajectory of the power transmission line; S5, allowing the drone after the corrected cruising path to photograph the power transmission line to obtain a corrected transmission line photo; wherein the power transmission line has one or more, and when there is only one power transmission line, the camera only collects the corrected transmission line photo of the one power transmission line; When there are multiple power transmission lines, the cameras are provided in plurality, and the multiple cameras correspond to the multiple power transmission lines one by one, the initial transmission line photo is taken by one of the cameras, and the revised transmission line photo includes photos taken by the multiple cameras; S6. Based on computer vision recognition method, a foreign body recognition model for transmission line is constructed; S7, identifying defects of the power transmission line by inputting the corrected transmission line photo into the transmission line foreign body recognition model to obtain a recognition result; S8. Determine whether the power transmission line is abnormal based on the identification result.

2. The power monitoring method based on drone cruising according to claim 1 is characterized in that: The initial UAV cruising path is created based on the location information of the power transmission lines, which includes the following steps: S101, obtaining the position coordinates and altitudes of both ends of the power transmission line; S102, creating an initial flight instruction according to the position coordinates and altitude of both ends of the power transmission line; S103: construct the initial cruise path using the initial flight instruction.

3. The power monitoring method based on drone cruising according to claim 1 is characterized in that: Allowing the drone to cruise along the initial cruise path to photograph the power transmission line to obtain an initial transmission line photo specifically includes the following steps: S201, allowing the UAV to cruise along an initial cruise path, so that the UAV is located above the power transmission line for cruising; S202, allowing the drone to take photos vertically downward while cruising above the power transmission line to obtain the initial transmission line photo.

4. The power monitoring method based on drone cruising according to claim 1 is characterized in that: By identifying the position of the power transmission line on the initial transmission line photo, obtaining the cruising path deviation value of the UAV specifically includes the following steps: S301, establishing a three-dimensional rectangular coordinate system with the focus center of the camera as the origin, the optical axis of the camera as the Z axis, and the motion direction of the drone as the X axis, to obtain a camera coordinate system; S302, establishing a three-dimensional rectangular coordinate system with the intersection of the optical axis of the camera and the image plane of the initial transmission line photo as the coordinate origin, the optical axis of the camera as the Z axis, and the projection of the X axis of the camera coordinate system on the initial transmission line photo as the X axis to obtain an image coordinate system; S303, calculating the Y coordinate value of the center point of the power transmission line in the initial transmission line photo passing through the Y axis of the image coordinate system in the image coordinate system; S304, using the external parameter matrix of the camera to convert the Y coordinate value of the center point of the power transmission line in the image coordinate system to the camera coordinate system, to obtain the cruise path deviation value.

5. The power monitoring method based on drone cruising according to claim 4 is characterized in that: Calculating the Y coordinate value of the center point of the power transmission line in the initial transmission line photo passing through the Y axis of the image coordinate system in the image coordinate system specifically includes the following steps: The initial transmission line photo is subjected to grayscale processing to obtain an initial transmission line grayscale image; Acquire Y coordinate values ​​of a plurality of pixel points on the Y axis of the image coordinate system in the initial transmission line grayscale image to obtain a plurality of pixel Y coordinate values; Binarize each pixel point on the Y axis of the image coordinate system to obtain a plurality of corresponding pixel point grayscale values; Calculate the difference between two pixel points whose grayscale values ​​are within the preset range in the Y coordinate direction to obtain the central pixel difference; The sum of the central pixel difference and the Y coordinate value of the near origin is calculated to obtain the Y coordinate value of the center point of the power transmission line in the image coordinate system; wherein the near origin is a pixel point close to the origin of the image coordinate system among the two pixel points whose grayscale values ​​are within a preset range.

6. The power monitoring method based on drone cruising according to claim 4 is characterized in that: Correcting the cruise path of the UAV according to the cruise path deviation value specifically includes the following steps: Using the GPS navigation system or the Beidou navigation system to construct the position coordinate system of the UAV; The cruise path deviation value is converted into the position coordinate system using the camera conversion matrix of the drone to obtain a navigation deviation value; The cruising path of the UAV is corrected according to the navigation deviation value.

7. The power monitoring method based on drone cruising according to claim 1 is characterized in that: Based on computer vision recognition method, a foreign body recognition model for transmission line is constructed, which includes the following steps: S601. Constructing an initial visual recognition model based on computer visual recognition methods; S602, collecting a plurality of abnormal images of the power transmission line to obtain a plurality of initial training images; S603, preprocessing the multiple initial training images to remove noise in each of the initial training images to obtain multiple formal training images; S604: Use the plurality of formal training images to train the initial visual recognition model to obtain the transmission line foreign object recognition model.

8. The power monitoring method based on drone cruising according to claim 1 is characterized in that: The defects of the power transmission line are identified by inputting the corrected transmission line photo into the transmission line foreign body identification model to obtain an identification result, which specifically includes the following steps: S701, inputting the modified transmission line photo into the transmission line foreign body recognition model, so as to recognize feature vectors in the modified transmission line photo through the transmission line foreign body recognition model, and obtain multiple feature vectors; S702. Calculate the feature matching degree between each of the feature vectors and the plurality of abnormal feature vectors respectively using an approximate nearest neighbor matching method to obtain a plurality of feature matching degree values; wherein the recognition result includes the plurality of feature matching degree values.

9. The power monitoring method based on drone cruising according to claim 8 is characterized in that: Judging whether the power transmission line is abnormal according to the recognition result, the specific steps are: Each of the feature matching values ​​is compared with a preset threshold. If any of the feature matching values ​​is greater than or equal to the preset threshold, the power transmission line is judged to be abnormal; if all of the feature matching values ​​are less than the preset threshold, the power transmission line is judged to be normal.

10. An electric power monitoring system based on drone cruising, characterized in that: The power transmission line is monitored by using the power monitoring method based on drone cruising as claimed in any one of claims 1 to 9, comprising: A controller, configured to create an initial cruise path according to the location information of the power transmission line, and to create an initial control instruction according to the initial cruise path; A drone is connected to the controller through wireless communication, and is used to cruise the power transmission line under the control of the initial control instruction to photograph the power transmission line to obtain an initial transmission line photo; A data processing server establishes a communication connection with the controller, and is used to obtain the initial power transmission line photo through the controller; obtain the cruise path deviation value of the UAV by identifying the position of the power transmission line on the initial power transmission line photo; correct the cruise path of the UAV according to the cruise path deviation value, and transmit the corrected cruise path of the UAV to the controller; The controller is also used to generate a correction control instruction according to the corrected cruising path of the UAV; The drone is also used to cruise along the extension trajectory of the power transmission line under the control of the revised control instruction to photograph the power transmission line and obtain a revised transmission line photo; The data processing server is also used to obtain the corrected transmission line photo through the controller; construct a transmission line foreign object recognition model based on a computer vision recognition method; identify the defects of the power transmission line by inputting the corrected transmission line photo into the transmission line foreign object recognition model to obtain a recognition result; and determine whether the power transmission line is abnormal based on the recognition result.

Images