A power cable inspection method and system based on drone

Abstract

The present invention is applicable to the field of electric power inspection technology, and in particular, relates to a method and system for electric power cable inspection based on a drone, the method comprising: constructing an electric power cable layout model; generating an electric power inspection route, setting a cable inspection point on the electric power inspection route, collecting images at the cable inspection point, and obtaining an original inspection image; processing the original inspection image to obtain a preprocessed image, and extracting a line diagram of electric power cables; transmitting the line diagram of electric power cables to a server, generating an electric power cable inspection model in an electric power cable layout model based on the line diagram of electric power cables, and determining whether there is an abnormality based on the electric power cable inspection model. The present invention preprocesses the collected image, and only transmits the line diagram where the cable is located, thereby restoring the cable image using the electric power cable layout model, and performing abnormality identification based on the restored image, which greatly reduces the amount of data transmission, and can provide stable data transmission even in areas with poor signal conditions.

Description

Technical Field

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

[0002] Power cables are conductor materials used to transmit electrical energy. They are usually made of metals with good electrical conductivity (such as copper or aluminum) and are wrapped with an insulating layer to prevent current leakage and ensure safe use. They are designed with different specifications and types according to their different uses. For example, high-voltage transmission cables are used to transmit large amounts of electrical energy over long distances, while low-voltage distribution cables are commonly used for power distribution inside homes and industrial buildings. The quality of power cables directly affects the stability and safety of the power system, so relevant standards and specifications must be strictly followed during selection and installation.

[0003] In the current use of power cables, manual inspection is generally adopted. With the development of technology, drone technology is gradually applied in the inspection process of power cables. However, drones need to transmit images in real time, and the data transmission volume is large, which affects the flight time of drones. Summary of the invention

[0004] The purpose of the present invention is to provide a power cable inspection method based on a drone, aiming to solve the problem that the drone needs to perform real-time image transmission, the data transmission volume is large, and the flight time of the drone is affected.

[0005] The present invention is implemented as follows: a power cable inspection method based on a drone, the method comprising:

[0006] Constructing a power cable layout model, wherein the power cable layout model includes tower information and cable information;

[0007] Generate a power inspection route based on the power cable layout model, set cable inspection points on the power inspection route, collect images at the cable inspection points, and obtain the original inspection image;

[0008] Processing the original inspection image to obtain a preprocessed image, extracting a power cable line diagram based on the preprocessed image, wherein the number of the power cable line diagrams is two groups, and the shooting positions of the original inspection images corresponding to the two groups of power cable line diagrams are different;

[0009] The power cable line diagram is transmitted to the server, a power cable inspection model is generated in the power cable layout model based on the power cable line diagram, and whether there is an abnormality is determined based on the power cable inspection model.

[0010] Preferably, the steps of generating a power inspection route based on the power cable layout model, setting cable inspection points on the power inspection route, and collecting images at the cable inspection points to obtain the original inspection image specifically include:

[0011] Read the cable information in the power cable layout model, determine the moving path of the drone according to the layout position of the cables to be inspected, and obtain the power inspection path;

[0012] Extract the positions of all towers on the power inspection path, and determine at least two cable inspection points between adjacent towers based on the distances between adjacent towers;

[0013] The drone is controlled to fly along the power inspection route. When the drone reaches the nearest position to the cable inspection point, it goes to the cable inspection point to collect images of the power cables. When collecting images, the drone's location information and image collection parameters are recorded.

[0014] Preferably, the step of processing the original inspection image to obtain a preprocessed image and extracting the power cable line diagram based on the preprocessed image specifically includes:

[0015] Perform grayscale processing on the original inspection image, convert it into a grayscale image, adjust the contrast of the grayscale image, and obtain a preprocessed image;

[0016] Perform pixel statistics on the preprocessed image, divide the pixels into regions based on a preset gray value range and pixel position relationship, and obtain multiple pixel connected domains;

[0017] The minimum circumscribed ellipse is generated based on the pixel connected domain, and the corresponding pixel connected domain is extracted according to the ratio of the major axis to the minor axis of the minimum circumscribed ellipse to generate the power cable line diagram.

[0018] Preferably, the steps of transmitting the power cable line diagram to the server, generating a power cable inspection model in the power cable layout model based on the power cable line diagram, and determining whether there is an abnormality based on the power cable inspection model specifically include:

[0019] The power cable line diagram is transmitted to the server, and the location information and image acquisition parameters of the corresponding drone are received through the server;

[0020] A virtual camera is constructed in the power cable layout model, and a radial model is constructed according to two sets of power cable line drawings. The intersection of the two sets of radial models is the power cable inspection model.

[0021] Calculate the real-time parameter information of the power cable inspection model and compare it with the original cable information to determine whether there is any abnormality.

[0022] Preferably, the abnormal situation of the cable includes that the cable length change rate exceeds a preset value, the cable sag exceeds a preset value, and the cable quantity does not match.

[0023] Another object of the present invention is to provide a power cable inspection system based on a drone, the system comprising:

[0024] A laying model building module is used to build a power cable laying model, wherein the power cable laying model includes tower information and cable information;

[0025] An image acquisition module is used to generate a power inspection route based on a power cable layout model, set cable inspection points on the power inspection route, and perform image acquisition at the cable inspection points to obtain an original inspection image;

[0026] A cable image extraction module is used to process the original inspection image to obtain a preprocessed image, and extract a power cable line diagram based on the preprocessed image. The number of the power cable line diagrams is two groups, and the shooting positions of the original inspection images corresponding to the two groups of power cable line diagrams are different;

[0027] The abnormality identification module is used to transmit the power cable line diagram to the server, generate a power cable inspection model in the power cable layout model based on the power cable line diagram, and determine whether there is an abnormality based on the power cable inspection model.

[0028] Preferably, the image acquisition module includes:

[0029] The inspection path generation unit is used to read the cable information in the power cable layout model, determine the moving path of the drone according to the layout position of the cable to be inspected, and obtain the power inspection path;

[0030] A patrol point setting unit, used to extract the positions of all towers on the power patrol path, and determine at least two cable patrol points between adjacent towers based on the distance between adjacent towers;

[0031] The route inspection unit is used to control the drone to fly along the power inspection route. When the drone reaches the nearest position to the cable inspection point, it goes to the cable inspection point to collect images of the power cables. When collecting images, the drone's location information and image collection parameters are recorded.

[0032] Preferably, the cable image extraction module includes:

[0033] The image preprocessing unit is used to perform grayscale processing on the original inspection image, convert it into a grayscale image, adjust the contrast of the grayscale image, and obtain a preprocessed image;

[0034] A pixel statistics unit is used to perform pixel statistics on the preprocessed image, divide the pixels into regions based on a preset gray value range and pixel position relationship, and obtain multiple pixel connected domains;

[0035] The line diagram generating unit is used to generate a minimum circumscribed ellipse based on a pixel connected domain, extract the corresponding pixel connected domain according to the ratio of the major axis to the minor axis of the minimum circumscribed ellipse, and generate a power cable line diagram.

[0036] Preferably, the abnormality identification module includes:

[0037] A data transmission unit, used to transmit the power cable line diagram to the server, and receive the corresponding UAV position information and image acquisition parameters through the server;

[0038] A model building unit is used to build a virtual camera in the power cable layout model, and to build a radial model according to two sets of power cable line drawings. The intersection of the two sets of radial models is the power cable inspection model.

[0039] The abnormality determination unit is used to calculate the real-time parameter information of the power cable inspection model, compare it with the original cable information, and determine whether there is an abnormality.

[0040] Preferably, the abnormal situation of the cable includes that the cable length change rate exceeds a preset value, the cable sag exceeds a preset value, and the cable quantity does not match.

[0041] The present invention provides a power cable inspection method based on a drone. The drone is used to collect images of power cables, and the collected images are preprocessed to transmit only the line drawings where the cables are located. The cable images are then restored using a power cable layout model, and abnormalities are identified based on the restored images, thereby greatly reducing the amount of data transmission and providing stable data transmission even in areas with poor signal conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A flowchart of a power cable inspection method based on a drone provided in an embodiment of the present invention;

[0043] Figure 2 A flowchart of the steps of generating a power inspection route based on a power cable layout model, setting cable inspection points on the power inspection route, collecting images at the cable inspection points, and obtaining an original inspection image provided by an embodiment of the present invention;

[0044] Figure 3 A flowchart of the steps of processing an original inspection image to obtain a preprocessed image and extracting a power cable line diagram based on the preprocessed image provided by an embodiment of the present invention;

[0045] Figure 4 A flowchart of the steps of transmitting a power cable line diagram to a server, generating a power cable inspection model in a power cable layout model based on the power cable line diagram, and determining whether there is an abnormality based on the power cable inspection model provided in an embodiment of the present invention;

[0046] Figure 5 An architecture diagram of a power cable inspection system based on a drone provided in an embodiment of the present invention;

[0047] Figure 6 An architecture diagram of an image acquisition module provided by an embodiment of the present invention;

[0048] Figure 7 An architecture diagram of a cable image extraction module provided by an embodiment of the present invention;

[0049] Figure 8 An architecture diagram of an abnormality identification module provided by an embodiment of the present invention;

[0050] Fig. 9 A schematic diagram of the principle of generating a power cable inspection model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0052] like Figure 1 FIG. 1 is a flowchart of a power cable inspection method based on a drone provided by an embodiment of the present invention, the method comprising:

[0053] S100: construct a power cable layout model, where the power cable layout model includes tower information and cable information.

[0054] In this step, a power cable laying model is constructed. In the power cable laying model, the location information of all towers is recorded. The tower information also includes the height and style of the towers. Specifically, a corresponding model is constructed in the power cable laying model. The power cable laying model is a three-dimensional model. The cables installed between the towers are recorded through cable information. The corresponding model is also constructed according to the length, verticality and other information of the cables. Then, based on the power cable laying model, the position coordinates of any point on the tower and the length information of the cable can be determined.

[0055] S200, generating a power inspection route based on a power cable layout model, setting cable inspection points on the power inspection route, and collecting images at the cable inspection points to obtain original inspection images.

[0056] In this step, a power inspection path is generated based on the power cable layout model. The cables are continuous. When a certain part of the cables needs to be inspected, the corresponding cables are selected. Based on the installation position of the cables, the corresponding power inspection path will be automatically generated according to the layout position and direction of the cables. Specifically, the power inspection path can be set at a height of 20 meters directly above the cables so that it is parallel to the cables that need to be inspected. When inspecting the cables, the main focus is on inspecting whether there are any abnormalities in the cables between the two towers. Therefore, cable inspection points are set between two groups of adjacent towers. The number of cable inspection points is at least two groups. When the drone arrives near the cable inspection point, it flies to the cable inspection point to collect images and obtain the original inspection image.

[0057] S300, processing the original inspection image to obtain a preprocessed image, and extracting a power cable line diagram based on the preprocessed image, wherein the number of the power cable line diagrams is two groups, and the shooting positions of the original inspection images corresponding to the two groups of power cable line diagrams are different.

[0058] In this step, the original inspection image is processed, and the image acquisition device on the drone acquires the original inspection image. If it is directly transmitted in real time, the data transmission volume is very large. Many towers are built in uninhabited areas, and the signal strength of mobile communications is low. It is difficult to complete the transmission of a large amount of data in a short time. If real-time transmission is not performed, the staff cannot know the situation on the scene in real time. Therefore, the drone processes the original inspection image to obtain a preprocessed image, processes it into a grayscale image, and extracts the power cable line diagram through pixel statistics. During transmission, only the power cable line diagram is transmitted, and other pixels are not transmitted, and only a single grayscale value is used to characterize the pixels on the power cable, which greatly reduces the data transmission volume and ensures the timeliness of data transmission.

[0059] S400, transmitting the power cable line diagram to the server, generating a power cable inspection model in the power cable layout model based on the power cable line diagram, and determining whether there is an abnormality based on the power cable inspection model.

[0060] In this step, the power cable line diagram is transmitted to the server. The power cable line diagram contains a very small amount of data and can be transmitted to the server in real time. The power cable layout model is loaded in the server, and the corresponding virtual camera is generated in the power cable layout model based on the acquisition parameters of the inspection original image. The power cable line diagram is projected onto the acquisition screen of the virtual camera, thereby generating a power cable inspection model in the power cable layout model. The model of the power cable is determined based on this, that is, the power cable inspection model. Subsequently, the current state of the power cable is compared with the preset state of the power cable to determine whether there is an abnormality. When it is determined that there is an abnormality, the drone is controlled to transmit the corresponding inspection original image to provide a clearer real-time image.

[0061] like Figure 2 As shown, as a preferred embodiment of the present invention, the steps of generating a power inspection path based on a power cable layout model, setting a cable inspection point on the power inspection path, and performing image acquisition at the cable inspection point to obtain an original inspection image specifically include:

[0062] S201, reading the cable information in the power cable layout model, determining the moving path of the drone according to the layout positions of the cables to be inspected, and obtaining the power inspection path.

[0063] In this step, the cable information in the power cable laying model is read. The power cable laying model records the initial laying status of the cable, including length, span, wire diameter, etc. The coordinates of each point on the power cable can be known, so a power inspection path parallel to the power cable can be set according to the position of the power cable.

[0064] S202, extracting the positions of all towers on the power inspection path, and determining at least two cable inspection points between adjacent towers based on the distances between adjacent towers.

[0065] In this step, the positions of all towers on the power inspection path are extracted. Since the power cables are laid between two groups of towers, when the power cables are abnormal, such as abnormal contraction causing the cables to be tightened, resulting in excessive tension on the cables, affecting the service life of the cables, the properties of the power cables such as sag and length will change. In order to collect the above images, at least two groups of cable inspection points are set between adjacent towers to determine the positions of the power cables in the power cable laying model.

[0066] S203, controlling the drone to fly along the power inspection route. When the drone reaches the nearest position to the cable inspection point, it goes to the cable inspection point to collect images of the power cables. When collecting images, the drone's position information and image collection parameters are recorded.

[0067] In this step, the drone is controlled to fly along the power inspection route. When flying, the drone collects images of the power cables between any two groups of towers. In order to reduce the complexity of the picture, the cable inspection point is set below the power cable, and the image acquisition module of the drone collects images diagonally upward, that is, the sky is used as the background of the cable. On the one hand, it can ensure that the background is clean, and on the other hand, it can improve the contrast between the cable and the sky, so as to extract the image of the power cable.

[0068] like Figure 3As shown, as a preferred embodiment of the present invention, the steps of processing the original inspection image to obtain a preprocessed image and extracting the power cable line diagram based on the preprocessed image specifically include:

[0069] S301, grayscale processing is performed on the original inspection image to convert it into a grayscale image, and the contrast of the grayscale image is adjusted to obtain a preprocessed image.

[0070] In this step, the original inspection image is gray-processed. Gray-scale processing can greatly reduce the amount of data processing, and the gray-scale processing process is simple and requires low computing power. The contrast is then adjusted. By adjusting the contrast, the contrast between the power cable and the sky can be made higher, and a pre-processed image is obtained.

[0071] S302, performing pixel statistics on the preprocessed image, dividing the pixels into regions based on a preset gray value range and pixel position relationship, and obtaining a plurality of pixel connected domains.

[0072] In this step, pixel statistics are performed on the preprocessed image. Each time a pixel is selected, the selected pixel and its adjacent pixels are calculated, and the grayscale value difference between the two is calculated. If the grayscale value difference is lower than a preset value, it is divided into the same pixel connected domain, and the pixels included in the pixel connected domain are deemed to have been selected. In the subsequent statistical process, the selected pixels are no longer selected, and the selected pixels cannot be regarded as adjacent pixels. Based on this, the preprocessed image is divided into multiple pixel connected domains, and the part of the pixel connected domain that contains pixels exceeding the preset value in the longitudinal direction is eliminated. For example, a rectangular scanning area is set along the longitudinal direction (the transverse direction contains one pixel, and the longitudinal direction contains an infinite number of pixels), and the rectangular scanning area is moved along the transverse direction, one pixel at a time. If the adjacent pixels in the rectangular scanning area exceed the preset number, this part of the pixels is eliminated from the pixel connected domain.

[0073] S303, generating a minimum circumscribed ellipse based on the pixel connected domain, extracting the corresponding pixel connected domain according to the ratio of the major axis to the minor axis of the minimum circumscribed ellipse, and generating a power cable line diagram.

[0074] In this step, a minimum circumscribed ellipse is generated based on a pixel connected domain. If the angle between the major axis and the horizontal line is less than a preset value, and the ratio of the major axis to the minor axis is greater than a preset value, the pixel connected domain is retained as part of the power cable line diagram. The size of the preprocessed image is recorded, all pixels in the retained pixel connected domain are replaced with preset pixels, and the remaining pixels are set as blank pixels to obtain a power cable line diagram.

[0075] like Figure 4As shown, as a preferred embodiment of the present invention, the steps of transmitting the power cable line diagram to the server, generating a power cable inspection model in the power cable layout model based on the power cable line diagram, and determining whether there is an abnormality based on the power cable inspection model specifically include:

[0076] S401, transmitting the power cable line diagram to the server, and receiving the corresponding drone location information and image acquisition parameters through the server.

[0077] In this step, the power cable line diagram is transmitted to the server. The power cable layout model is stored in the server. When the server receives the power cable line diagram, it simultaneously receives the location information and image acquisition parameters of the drone. Based on the location information of the drone, the location of the power cables that the drone is currently inspecting can be determined.

[0078] S402, constructing a virtual camera in the power cable layout model, constructing a radial model according to the two groups of power cable line drawings, and the intersection of the two groups of radial models is the power cable inspection model.

[0079] In this step, a virtual camera is constructed in the power cable layout model. The parameters of the virtual camera and the image acquisition module on the drone are the same, and the screen ratio is the same. After the power cable layout model is loaded, the power cable line diagram is pasted in the acquisition screen of the virtual camera, and a set of radiation models is generated accordingly. This is because the image acquisition module is a monocular camera and cannot perceive the distance. When the volume and position of the object change, it can be ensured that its image in the image acquisition module does not change. In the present invention, the power cable line diagram has two groups to obtain two groups of radiation models. The intersection area of ​​the two groups of radiation models is the location of the power cable. The principle is as follows Fig. 9 shown.

[0080] S403, calculating the real-time parameter information of the power cable inspection model, comparing it with the original cable information, and determining whether there is an abnormality.

[0081] In this step, the real-time parameter information of the power cable inspection model is calculated, and the length, span, sag and other parameters that need to be verified of the power cables are counted and compared with the original data to determine whether the current power cables are abnormal; cable abnormalities include cable length change rate exceeding the preset value, cable sag exceeding the preset value and cable quantity mismatch. When an abnormality occurs, the drone is controlled to transmit the corresponding original inspection image to the server.

[0082] like Figure 5 As shown, a power cable inspection system based on a drone is provided in an embodiment of the present invention, and the system includes:

[0083] The laying model building module 100 is used to build a power cable laying model, wherein the power cable laying model includes tower information and cable information.

[0084] In this system, the laying model construction module 100 constructs a power cable laying model. In the power cable laying model, the location information of all towers is recorded. The tower information also includes the height and style of the towers. Specifically, a corresponding model is constructed in the power cable laying model. The power cable laying model is a three-dimensional model. The cables installed between the towers are recorded through cable information, and the corresponding model is also constructed according to the length, verticality and other information of the cables. Then, based on the power cable laying model, the position coordinates of any point on the tower and the length information of the cable can be determined.

[0085] The image acquisition module 200 is used to generate a power inspection route based on the power cable layout model, set cable inspection points on the power inspection route, and perform image acquisition at the cable inspection points to obtain an original inspection image.

[0086] In this system, the image acquisition module 200 generates a power inspection path based on the power cable layout model. The cables are continuous. When a certain part of the cables needs to be inspected, the corresponding cables are selected. Based on the installation position of the cables, the corresponding power inspection path will be automatically generated according to the layout position and direction of the cables. Specifically, the power inspection path can be set at a height of 20 meters directly above the cables so that it is parallel to the cables that need to be inspected. When inspecting the cables, the main focus is on inspecting whether there are any abnormalities in the cables between the two towers. Therefore, cable inspection points are set between two groups of adjacent towers. The number of cable inspection points is at least two groups. When the drone arrives near the cable inspection point, it flies to the cable inspection point to collect images and obtain the original inspection image.

[0087] The cable image extraction module 300 is used to process the original inspection image to obtain a preprocessed image, and extract the power cable line diagram based on the preprocessed image. The number of the power cable line diagrams is two groups, and the shooting positions of the original inspection images corresponding to the two groups of power cable line diagrams are different.

[0088] In this system, the cable image extraction module 300 processes the original inspection image, and the image acquisition device on the drone acquires the original inspection image. If real-time transmission is performed directly, the data transmission volume is very large. Many towers are built in uninhabited areas, and the signal strength of mobile communications is low. It is difficult to complete the transmission of a large amount of data in a short time. If real-time transmission is not performed, the staff cannot know the situation on the scene in real time. Therefore, the drone processes the original inspection image to obtain a preprocessed image, processes it into a grayscale image, and extracts the power cable line diagram through pixel statistics. During transmission, only the power cable line diagram is transmitted, and other pixels are not transmitted, and only a single grayscale value is used to characterize the pixels on the power cable, which greatly reduces the data transmission volume and ensures the timeliness of data transmission.

[0089] The abnormality identification module 400 is used to transmit the power cable line diagram to the server, generate a power cable inspection model in the power cable layout model based on the power cable line diagram, and determine whether there is an abnormality based on the power cable inspection model.

[0090] In this system, the abnormality identification module 400 transmits the power cable line diagram to the server. The power cable line diagram contains a very small amount of data and can be transmitted to the server in real time. The power cable laying model is loaded in the server, and the corresponding virtual camera is generated in the power cable laying model based on the acquisition parameters of the inspection original image. The power cable line diagram is projected onto the acquisition screen of the virtual camera, thereby generating a power cable inspection model in the power cable laying model. The model of the power cable is determined based on this, that is, the power cable inspection model. Subsequently, the current state of the power cable is compared with the preset state of the power cable to determine whether there is an abnormality. When it is determined that there is an abnormality, the drone is controlled to transmit the corresponding inspection original image to provide a clearer real-time image.

[0091] like Figure 6 As shown, as a preferred embodiment of the present invention, the image acquisition module 200 includes:

[0092] The inspection path generation unit 201 is used to read the cable information in the power cable layout model, determine the moving path of the drone according to the layout position of the cable to be inspected, and obtain the power inspection path.

[0093] In this module, the inspection path generation unit 201 reads the cable information in the power cable laying model. The power cable laying model records the initial laying status of the cable, including length, span, wire diameter, etc. The coordinates of each point on the power cable can be known, so a power inspection path parallel to the power cable can be set according to the position of the power cable.

[0094] The inspection point setting unit 202 is used to extract the positions of all towers on the power inspection path, and determine at least two cable inspection points between adjacent towers based on the distances between adjacent towers.

[0095] In this module, the inspection point setting unit 202 extracts the positions of all towers on the power inspection path. Since the power cables are laid between two groups of towers, when the power cables are abnormal, such as abnormal contraction causing the cables to be tightened, resulting in excessive tension on the cables, affecting the service life of the cables, the properties of the power cables such as sag and length will change. In order to collect the above images, at least two groups of cable inspection points are set between adjacent towers to facilitate determining the positions of the power cables in the power cable laying model.

[0096] The route inspection unit 203 is used to control the drone to fly along the power inspection route. When the drone reaches the nearest position to the cable inspection point, it goes to the cable inspection point to capture images of the power cables. When capturing images, the drone's location information and image capture parameters are recorded.

[0097] In this module, the route inspection unit 203 controls the drone to fly along the power inspection route. When flying, the drone collects images of the power cables between any two groups of towers. In order to reduce the complexity of the picture, the cable inspection point is set below the power cable. The image acquisition module of the drone collects images diagonally upward, that is, the sky is used as the background of the cable. On the one hand, it can ensure that the background is clean, and on the other hand, it can improve the contrast between the cable and the sky, so as to extract the image of the power cable.

[0098] like Figure 7 As shown, as a preferred embodiment of the present invention, the cable image extraction module 300 includes:

[0099] The image preprocessing unit 301 is used to perform grayscale processing on the original inspection image, convert it into a grayscale image, and adjust the contrast of the grayscale image to obtain a preprocessed image.

[0100] In this module, the image preprocessing unit 301 performs grayscale processing on the original inspection image. Grayscale processing can greatly reduce the amount of data processing, and the grayscale processing process is simple and requires low computing power. The contrast is then adjusted. By adjusting the contrast, the contrast between the power cable and the sky can be made higher, and a preprocessed image is obtained.

[0101] The pixel statistics unit 302 is used to perform pixel statistics on the preprocessed image, divide the pixels into regions based on a preset gray value range and pixel position relationship, and obtain multiple pixel connected domains.

[0102] In this module, the pixel statistics unit 302 performs pixel statistics on the pre-processed image, selects a pixel each time, calculates the selected pixel and its adjacent pixels, and calculates the grayscale value difference between the two. If the grayscale value difference is lower than a preset value, it is divided into the same pixel connected domain, and the pixels included in the pixel connected domain are deemed to have been selected. In the subsequent statistical process, the selected pixels are no longer selected, and the selected pixels cannot be regarded as adjacent pixels. Based on this, the pre-processed image is divided into multiple pixel connected domains, and the part of the pixel connected domain that contains pixels exceeding the preset value in the longitudinal direction is eliminated. For example, a rectangular scanning area is set along the longitudinal direction (the transverse direction contains one pixel, and the longitudinal direction contains an infinite number of pixels), and the rectangular scanning area is moved along the transverse direction, one pixel at a time. If the adjacent pixels in the rectangular scanning area exceed the preset number, this part of the pixels is eliminated from the pixel connected domain.

[0103] The line diagram generating unit 303 is used to generate a minimum circumscribed ellipse based on the pixel connected domain, extract the corresponding pixel connected domain according to the ratio of the major axis to the minor axis of the minimum circumscribed ellipse, and generate a power cable line diagram.

[0104] In this module, the line diagram generation unit 303 generates a minimum circumscribed ellipse based on the pixel connected domain. If the angle between the major axis and the horizontal line is less than a preset value, and the ratio of the major axis to the minor axis is greater than a preset value, the pixel connected domain is retained as part of the power cable line diagram, the size of the preprocessed image is recorded, all pixels in the retained pixel connected domain are replaced with preset pixels, and the remaining pixels are set as blank pixels to obtain the power cable line diagram.

[0105] like Figure 8 As shown, as a preferred embodiment of the present invention, the abnormality identification module 400 includes:

[0106] The data transmission unit 401 is used to transmit the power cable line diagram to the server, and receive the corresponding drone position information and image acquisition parameters through the server.

[0107] In this module, the data transmission unit 401 transmits the power cable line diagram to the server. The power cable layout model is stored in the server. When the server receives the power cable line diagram, it simultaneously receives the location information and image acquisition parameters of the drone. Based on the location information of the drone, the location of the power cables that the drone is currently inspecting can be determined.

[0108] The model building unit 402 is used to build a virtual camera in the power cable layout model, build a radial model according to the two groups of power cable line drawings, and the intersection of the two groups of radial models is the power cable inspection model.

[0109] In this module, the model building unit 402 constructs a virtual camera in the power cable laying model. The virtual camera has the same parameters as the image acquisition module on the drone and has the same screen ratio. After loading the power cable laying model, the power cable line diagram is pasted in the acquisition screen of the virtual camera, and a set of radiation models is generated accordingly. This is because the image acquisition module is a monocular camera and cannot perceive the distance. When the volume and position of the object change, it can be ensured that its image in the image acquisition module does not change. In the present invention, the power cable line diagram has two groups to obtain two groups of radiation models. The intersection area of ​​the two groups of radiation models is the location of the power cable. The principle is as follows Fig. 9 shown.

[0110] The abnormality determination unit 403 is used to calculate the real-time parameter information of the power cable inspection model, compare it with the original cable information, and determine whether there is an abnormality.

[0111] In this module, the abnormality judgment unit 403 calculates the real-time parameter information of the power cable inspection model, counts the length, span, sag and other parameters that need to be verified of the power cable, and compares them with the original data to determine whether the current power cable has an abnormality; cable abnormalities include the cable length change rate exceeding the preset value, the cable sag exceeding the preset value and the cable quantity mismatch. When an abnormality occurs, the drone is controlled to transmit the corresponding original inspection image to the server.

[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 and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A power cable inspection method based on drone, characterized in that: The method comprises: Constructing a power cable layout model, wherein the power cable layout model includes tower information and cable information; Generate a power inspection route based on the power cable layout model, set cable inspection points on the power inspection route, collect images at the cable inspection points, and obtain the original inspection image; Processing the original inspection image to obtain a preprocessed image, extracting a power cable line diagram based on the preprocessed image, wherein the number of the power cable line diagrams is two groups, and the shooting positions of the original inspection images corresponding to the two groups of power cable line diagrams are different; The power cable line diagram is transmitted to the server, a power cable inspection model is generated in the power cable layout model based on the power cable line diagram, and whether there is an abnormality is determined based on the power cable inspection model; The step of processing the original inspection image to obtain a preprocessed image and extracting the power cable line diagram based on the preprocessed image specifically includes: Perform grayscale processing on the original inspection image, convert it into a grayscale image, adjust the contrast of the grayscale image, and obtain a preprocessed image; Perform pixel statistics on the preprocessed image, divide the pixels into regions based on a preset gray value range and pixel position relationship, and obtain multiple pixel connected domains; The minimum circumscribed ellipse is generated based on the pixel connected domain, and the corresponding pixel connected domain is extracted according to the ratio of the major axis to the minor axis of the minimum circumscribed ellipse to generate the power cable line diagram.

2. The power cable inspection method based on drone according to claim 1 is characterized in that: The steps of generating a power inspection route based on the power cable layout model, setting cable inspection points on the power inspection route, and collecting images at the cable inspection points to obtain the original inspection images specifically include: Read the cable information in the power cable layout model, determine the moving path of the drone according to the layout position of the cables to be inspected, and obtain the power inspection path; Extract the positions of all towers on the power inspection path, and determine at least two cable inspection points between adjacent towers based on the distances between adjacent towers; The drone is controlled to fly along the power inspection route. When the drone reaches the nearest position to the cable inspection point, it goes to the cable inspection point to collect images of the power cables. When collecting images, the drone's location information and image collection parameters are recorded.

3. The power cable inspection method based on drone according to claim 2 is characterized in that: The steps of transmitting the power cable line diagram to the server, generating a power cable inspection model in the power cable layout model based on the power cable line diagram, and determining whether there is an abnormality based on the power cable inspection model specifically include: The power cable line diagram is transmitted to the server, and the location information and image acquisition parameters of the corresponding drone are received through the server; A virtual camera is constructed in the power cable layout model, and a radial model is constructed according to two sets of power cable line drawings. The intersection of the two sets of radial models is the power cable inspection model. Calculate the real-time parameter information of the power cable inspection model and compare it with the original cable information to determine whether there is any abnormality.

4. The power cable inspection method based on drone according to claim 1, characterized in that: Abnormal cable conditions include cable length change rate exceeding a preset value, cable sag exceeding a preset value, and cable quantity mismatch.

5. A power cable inspection system based on drone, characterized in that: The system comprises: A laying model building module is used to build a power cable laying model, wherein the power cable laying model includes tower information and cable information; An image acquisition module is used to generate a power inspection route based on a power cable layout model, set cable inspection points on the power inspection route, and perform image acquisition at the cable inspection points to obtain an original inspection image; A cable image extraction module is used to process the original inspection image to obtain a preprocessed image, and extract a power cable line diagram based on the preprocessed image. The number of the power cable line diagrams is two groups, and the shooting positions of the original inspection images corresponding to the two groups of power cable line diagrams are different; An abnormality identification module is used to transmit the power cable line diagram to the server, generate a power cable inspection model in the power cable layout model based on the power cable line diagram, and determine whether there is an abnormality based on the power cable inspection model; The cable image extraction module comprises: The image preprocessing unit is used to perform grayscale processing on the original inspection image, convert it into a grayscale image, adjust the contrast of the grayscale image, and obtain a preprocessed image; A pixel statistics unit is used to perform pixel statistics on the preprocessed image, divide the pixels into regions based on a preset gray value range and pixel position relationship, and obtain multiple pixel connected domains; The line diagram generating unit is used to generate a minimum circumscribed ellipse based on a pixel connected domain, extract the corresponding pixel connected domain according to the ratio of the major axis to the minor axis of the minimum circumscribed ellipse, and generate a power cable line diagram.

6. The power cable inspection system based on drone according to claim 5 is characterized in that: The image acquisition module comprises: The inspection path generation unit is used to read the cable information in the power cable layout model, determine the moving path of the drone according to the layout position of the cable to be inspected, and obtain the power inspection path; A patrol point setting unit, used to extract the positions of all towers on the power patrol path, and determine at least two cable patrol points between adjacent towers based on the distance between adjacent towers; The route inspection unit is used to control the drone to fly along the power inspection route. When the drone reaches the nearest position to the cable inspection point, it goes to the cable inspection point to collect images of the power cables. When collecting images, the drone's location information and image collection parameters are recorded.

7. The power cable inspection system based on drone according to claim 6 is characterized in that: The abnormality identification module comprises: A data transmission unit, used to transmit the power cable line diagram to the server, and receive the corresponding UAV position information and image acquisition parameters through the server; A model building unit is used to build a virtual camera in the power cable layout model, and to build a radial model according to two sets of power cable line drawings. The intersection of the two sets of radial models is the power cable inspection model. The abnormality determination unit is used to calculate the real-time parameter information of the power cable inspection model, compare it with the original cable information, and determine whether there is an abnormality.

8. The power cable inspection system based on drone according to claim 5, characterized in that: Abnormal cable conditions include cable length change rate exceeding a preset value, cable sag exceeding a preset value, and cable quantity mismatch.

Images