CN116245844A - Intelligent distribution line defect identification method and system based on vision multi-mode fusion - Google Patents
Intelligent distribution line defect identification method and system based on vision multi-mode fusion Download PDFInfo
- Publication number
- CN116245844A CN116245844A CN202310216882.4A CN202310216882A CN116245844A CN 116245844 A CN116245844 A CN 116245844A CN 202310216882 A CN202310216882 A CN 202310216882A CN 116245844 A CN116245844 A CN 116245844A
- Authority
- CN
- China
- Prior art keywords
- module
- defects
- unmanned aerial
- aerial vehicle
- infrared
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention provides a distribution line defect intelligent identification method and system based on vision multi-mode fusion, and belongs to the technical field of distribution line defect identification; the problem of low inspection defect discovery rate of the existing unmanned aerial vehicle is solved; the method comprises the following steps: collecting pictures of an infrared camera and a zoom camera under different far and near focal lengths through an unmanned aerial vehicle; carrying out affine matrix solution on each group of acquired infrared images and RGB images, and solving affine matrixes from the RGB images to the infrared images; training all acquired RGB images by using a yolov5l target detection model; cutting and separating infrared and visible light spliced video streams captured by an unmanned aerial vehicle, transmitting the visible light video streams to a trained yolov5l target detection model for detection, judging a detected target, directly marking the detected target in an image if the detected target is a component defect, mapping the detected target into an infrared image through an affine matrix if the detected target is the component target, and judging whether the component has a thermal defect or not; the method is applied to line defect identification.
Description
Technical Field
The invention provides a distribution line defect intelligent identification method and system based on vision multi-mode fusion, and belongs to the technical field of distribution line defect identification.
Background
Along with the deep advancement of digital construction of the national network, the construction of a novel clean, digital and intelligent power system becomes the development direction of transformation of the power industry, deepens state sensing and analysis processing on the power grid and equipment, and has a vital influence on guaranteeing the reliability and safety of the power system. The distribution network plays a very important role in the production and operation of electric power as a main transmission system for connecting power supply equipment and power utilization customers, and the safety and reliability of the distribution network directly determine the functions and the efficiency of the power supply equipment and the power utilization equipment. At present, the operation and maintenance team of the power distribution network still adopts a large amount of manual modes to patrol in the line patrol work, mainly by electric workers turning mountain and going over the mountain, adopts manual observation or unmanned aerial vehicle image to check and patrol the existing power line, and has the defects of high operation intensity, long operation period, non-visual data, low precision, low reuse degree, difficult work in terrain complex areas and the like.
In recent years, national power grid companies actively push helicopters, unmanned aerial vehicles and manual collaborative inspection work, and the operation and maintenance levels of distribution lines are continuously improved. The unmanned aerial vehicle inspection operation reduces the safety risk brought by personnel climbing the tower, has the characteristics of high efficiency, good quality, small influence by the terrain condition and the like, and is an important means for the distribution line management to develop in a safer, efficient and fine direction. The number of defects found by inspection of the unmanned aerial vehicle is obviously increased, however, when the ground manual inspection and tower climbing maintenance are carried out, part of hardware defects are found to fail to be found by inspection of the unmanned aerial vehicle, so that the original inspection plan is changed, the workload is increased, and the working time is delayed. The defect efficiency of unmanned aerial vehicle inspection discovery is mainly caused by low positioning accuracy, low environment brightness and electromagnetic field interference.
Disclosure of Invention
The invention provides an intelligent distribution line defect identification method and system based on vision multi-mode fusion, which aim to solve the problem of low inspection defect discovery rate of the existing unmanned aerial vehicle.
In order to solve the technical problems, the invention adopts the following technical scheme: a distribution line defect intelligent identification method based on vision multi-mode fusion comprises the following steps:
s1: setting pictures acquired by the infrared camera and the zoom camera under different far and near focal lengths through the unmanned aerial vehicle, and recording focal length information of the infrared camera and the zoom camera and the corresponding acquired images;
s2: carrying out affine matrix solution on each group of acquired infrared images and RGB images, and solving affine matrixes from the RGB images to the infrared images;
s3: training all acquired RGB images by using a yolov5l target detection model, wherein two types of training are adopted, one type of training is a power transmission line component, the other type of training is a component defect, and the trained yolov5l target detection model is deployed on edge-end equipment carried by an unmanned aerial vehicle;
s4: cutting and separating infrared and visible light spliced video streams captured by an unmanned aerial vehicle, transmitting the visible light video streams to a yolov5l target detection model trained in edge equipment for detection, finally judging the detected target, directly marking the detected target in an image if the detected target is a component defect, mapping the detected target into an infrared image through an affine matrix if the detected target is a component target, solving the highest temperature and the lowest temperature in an infrared image marking frame, judging whether the component has a thermal defect or not, and marking the detected target in the image.
In the step S1, the two images acquired by the infrared camera and the zoom camera are images of the same size and position of the object acquired by the unmanned aerial vehicle at the same position and same height of the same component.
And in the step S3, a CBAM attention mechanism module is inserted into each CSP1_X module in the yolov5l basic network by the yolov5l target detection model, and important position information of a plurality of different scale feature maps is transmitted to the FPN+PAN network structure.
The network structure of the backbond part of the yolov5l target detection model comprises a first CBS module, a second CBS module, a first CSP1_3 module, a first CBAM module, a third CBS module, a CSP1_6 module, a second CBAM module, a fourth CBS module, a CSP1_9 module, a third CBAM module, a fifth CBS module, a second CSP1_3 module, a fourth CBAM module and an SPPF module which are sequentially connected.
The part defects comprise body defects and channel defects, wherein the body defects comprise joint heating, heat pins, missing nuts, insulator breakage, flashover, bird nests and floaters, and the channel defects are tree barriers.
The utility model provides a distribution line defect intelligent identification system based on vision multimode fuses, includes unmanned aerial vehicle and digital intelligence inspection management and control platform, be on-board on the unmanned aerial vehicle has infrared camera and zoom camera, marginal equipment, RTK positioning system, the deployment has distribution line defect intelligent identification method based on vision multimode fuses on the marginal equipment, be fixed with the LED light source on the fuselage of unmanned aerial vehicle.
The intelligent inspection management and control platform integrates an existing line library, a 3D point cloud library of a distribution network, an avionic point library, a sample library and a model library.
The zoom camera is provided with a 25mm fixed focus lens, and the aperture is F1.8.
The LED light source is fixed on the unmanned aerial vehicle body through a machine seat capable of being remotely controlled up and down and left and right, so that a dynamic light source is formed.
Compared with the prior art, the invention has the following beneficial effects: according to the intelligent identification method and system for the defects of the distribution line based on the vision multi-mode fusion, provided by the invention, the high-precision RTK positioning system, the customized 25mm long-focus lens and the additionally arranged LED light source are mounted on the unmanned aerial vehicle, so that the discovery rate of defects of the inspection hardware fittings of the unmanned aerial vehicle is improved, more hidden dangers can be discovered in the inspection of the line body, the line is kept at a healthy level, and the safe and stable operation of the line is ensured.
After the unmanned aerial vehicle is refitted, more hidden dangers can be found at one time in the line body inspection process, the problem that secondary inspection confirmation is carried out due to 'image blurring' and 'image darkness' is avoided, and therefore electricity consumption cost and vehicle outlet inspection times are reduced, and personnel cost and vehicle outlet cost are reduced.
The quality of the photo taken during inspection of the unmanned aerial vehicle is higher, six photos of the unmanned aerial vehicle version of the distribution line can be established, the unmanned aerial vehicle inspection photo library is established through the 10kV and above line towers in the jurisdiction, the unmanned aerial vehicle photo meeting the requirements is reserved, the basis is provided for later operation and maintenance and repair, and the number of manual inspection is reduced.
Drawings
The invention is further described below with reference to the accompanying drawings:
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a schematic diagram of a generic network architecture of the yolov5l model;
FIG. 3 is a schematic diagram of the basic network structure of the improved yolov5l object detection model of the present invention;
FIG. 4 is a flow chart of the invention for detecting an infrared image using a yolov5l object detection model;
FIG. 5 is a schematic diagram of the structure of the attention mechanism module of the CBAM of the present invention;
FIG. 6 is a schematic diagram of a channel attention module;
fig. 7 is a schematic structural diagram of the spatial attention module.
Detailed Description
As shown in fig. 1 to 7, the invention provides a distribution line defect intelligent identification method based on vision multi-mode fusion, which comprises the following steps:
s1: setting pictures acquired by the infrared camera and the zoom camera under different far and near focal lengths through the unmanned aerial vehicle, and recording focal length information of the infrared camera and the zoom camera and the corresponding acquired images;
s2: carrying out affine matrix solution on each group of acquired infrared images and RGB images, and solving affine matrixes from the RGB images to the infrared images;
s3: training all acquired RGB images by using a yolov5l target detection model, wherein two types of training are adopted, one type of training is a power transmission line component, the other type of training is a component defect, and the trained yolov5l target detection model is deployed on edge-end equipment carried by an unmanned aerial vehicle;
s4: cutting and separating infrared and visible light spliced video streams captured by an unmanned aerial vehicle, transmitting the visible light video streams to a yolov5l target detection model trained in edge equipment for detection, finally judging the detected target, directly marking the detected target in an image if the detected target is a component defect, mapping the detected target into an infrared image through an affine matrix if the detected target is a component target, solving the highest temperature and the lowest temperature in an infrared image marking frame, judging whether the component has a thermal defect or not, and marking the detected target in the image.
In the step S1, the two images acquired by the infrared camera and the zoom camera are images of the same size and position of the object acquired by the unmanned aerial vehicle at the same position and same height of the same component.
And in the step S3, a CBAM attention mechanism module is inserted into each CSP1_X module in the yolov5l basic network by the yolov5l target detection model, and important position information of a plurality of different scale feature maps is transmitted to the FPN+PAN network structure.
The network structure of the backbond part of the yolov5l target detection model comprises a first CBS module, a second CBS module, a first CSP1_3 module, a first CBAM module, a third CBS module, a CSP1_6 module, a second CBAM module, a fourth CBS module, a CSP1_9 module, a third CBAM module, a fifth CBS module, a second CSP1_3 module, a fourth CBAM module and an SPPF module which are sequentially connected.
The part defects comprise body defects and channel defects, wherein the body defects comprise joint heating, heat pins, missing nuts, insulator breakage, flashover, bird nests and floaters, and the channel defects are tree barriers.
The utility model provides a distribution line defect intelligent identification system based on vision multimode fuses, includes unmanned aerial vehicle and digital intelligence inspection management and control platform, be on-board on the unmanned aerial vehicle has infrared camera and zoom camera, marginal equipment, RTK positioning system, the deployment has distribution line defect intelligent identification method based on vision multimode fuses on the marginal equipment, be fixed with the LED light source on the fuselage of unmanned aerial vehicle.
The intelligent inspection management and control platform integrates an existing line library, a 3D point cloud library of a distribution network, an avionic point library, a sample library and a model library.
The zoom camera is provided with a 25mm fixed focus lens, and the aperture is F1.8.
The LED light source is fixed on the unmanned aerial vehicle body through a machine seat capable of being remotely controlled up and down and left and right, so that a dynamic light source is formed.
The invention provides a distribution line defect intelligent identification method based on vision multi-mode fusion, which mainly comprises the following steps:
1. firstly, setting pictures (the size and the position of objects in two images are approximately the same) acquired by an infrared camera and a zoom camera under a plurality of far and near focal lengths through an unmanned aerial vehicle, and recording focal length information of the infrared camera and the zoom camera and images acquired correspondingly;
2. carrying out affine matrix solution on each group of acquired infrared images and RGB images, and solving affine matrixes from the RGB images to the infrared images;
3. training all acquired RGB images by using a yolov5 target detection model, wherein two types of training are mainly adopted, one type of training is a power transmission line part, the other type of training is a part defect, and the trained model is deployed on edge-end equipment carried by an unmanned aerial vehicle;
4. cutting and separating infrared and visible light spliced video streams captured by an unmanned aerial vehicle, transmitting the visible light video streams to a trained target detection model in edge equipment for detection, finally judging the detected target, directly marking the detected target in an image if the detected target is a component defect (bird nest, floater, porcelain bottle damage and the like), mapping the detected target to an infrared image through an affine matrix if the detected target is a component target, solving the highest temperature and the lowest temperature in an infrared image marking frame, judging whether the component has a thermal defect or not, and marking the detected target in the image.
The invention adopts an improved yolov5l target detection model to detect defects in visible light images, the general yolov5l target detection model mainly comprises Backbone, neck and Prediction, the network structure is shown in figure 2 and mainly comprises CBS, CSP1_ X, SPPF, res unit and CSP2_X modules, conv is convolution operation, BN is normalization operation, siLU is an adopted activation function, add refers to the addition of values corresponding to feature images with the same size, concat refers to feature image splicing with the same size, maxPOOL refers to maximum pooling, and Unsample refers to up-sampling operation.
Some small component targets and defect targets on the distribution line are smaller, defect characteristics are difficult to extract, a missing detection phenomenon is easy to generate, and CBAM (Convolutional Block Attention Module, CBAM) is inserted into a foundation network of YOLOv5l to extract more abundant characteristic information according to the problem, and the structure is shown in figure 5.
The CBAM in turn contains two sub-modules, a channel attention module (Channel Attention Module, CAM) and a spatial attention module (Spatial Attention Module, SAM), respectively. As shown in fig. 6 and 7, the channel attention module may enable the network model to selectively focus on some important channels during the training and reasoning phase. The spatial attention module can weight the positions containing important features in the feature map, and further improves the detection of the target position by the model.
As shown in FIG. 3, the improved yolov5l target detection model of the invention inserts a CBAM attention mechanism module after each CSP1_X module in the yolov5l base network, because the feature images output by each CSP1_X module are inconsistent in size, after the CBAM is inserted, the important position information of a plurality of different scale feature images is transmitted to the FPN+PAN network structure, and the detection precision of distribution line components and defects thereof can be further improved.
Because the defects of the power distribution network equipment are various and complicated, the application requirements cannot be met by a single sensing equipment, and the method is used for completing the acquisition and labeling of the visible light and infrared multi-mode data based on important component defect types of the inspection attention of the 10 kilovolt power distribution network. Based on the marked trainable data set, the training of the transmission line defect identification model for the 10 kilovolt power distribution network is completed, and tests show that the real-time identification accuracy at the mobile end can reach 95% aiming at body defects such as joint heating, pin deficiency, nut deficiency, insulator breakage, flashover, bird nest, floater and channel defects such as tree barriers.
In order to detect the temperature of a target in an infrared image, but the target in an infrared image is difficult to detect directly through a yolov5 target detection model, the invention provides a method for detecting the infrared image based on a visible light image, which mainly comprises the following steps: (1) Calibrating an RGB camera and an infrared camera to obtain a matrix of mapping an RGB image to an infrared image; (2) Inputting the visible light image into a yolov5l target detection model to obtain a detection result; (3) And mapping the detection result to the infrared image through a mapping matrix, so as to obtain the specific position of the infrared image target. The detection schematic diagram is shown in fig. 4.
After knowing the target and its position contained in the infrared image, it can be used for making thermal defect diagnosis. The invention can realize the detection of infrared defects and visible light defects, and the detection results are shown in the following table 1:
table 1 defect detection results.
The invention also provides a distribution line defect intelligent recognition system based on vision multi-mode fusion, which is mainly used for solving the problems of low positioning accuracy, low environment brightness, fuzzy acquired line target image, darker image and the like caused by electromagnetic field interference in the existing unmanned aerial vehicle line defect detection, improving the existing unmanned aerial vehicle, increasing a high-accuracy RTK positioning system, realizing accurate positioning within a range of 10 cm, and having high single-tower inspection efficiency and high defect discovery rate. Install the LED light source additional at unmanned aerial vehicle fuselage and carry out the light filling, directly tie up on the fuselage undercarriage, do not destroy original unmanned aerial vehicle inside electrical structure and fuselage auxiliary equipment, along with unmanned aerial vehicle together aerial illumination, shoot object luminance and promote obviously, and the low power dissipation, with low costs, it is convenient to install. And the LED light source can be arranged on a base which can be remotely controlled up, down, left and right to form a dynamic light source. The customized 25mm long-focus lens is adopted, the picture quality is improved obviously, the electric pole part can still be seen clearly after 6 times of amplification, the imaging effect is good, the picture resolution is improved greatly, a device can be shot remotely, and the device is not required to be close to an electrified wire, so that electromagnetic interference is avoided.
The intelligent distribution line defect identification system based on vision multi-mode fusion also comprises a digital intelligent patrol management and control platform for ground monitoring, wherein the platform integrates basic data such as an existing line library and the like, and combines data such as a distribution network 3D point cloud library, a navigation point library, a sample library, a model library and the like to construct a first distribution network line digital warehouse, and has core functions such as intelligent patrol equipment management, distribution network equipment management, route management, task management, remote management and control, channel service and the like.
The intelligent distribution line defect recognition system based on vision multi-mode fusion provided by the invention forms a routing inspection flow of 3D path planning, autonomous flight, intelligent focusing, "infrared and visible light" intelligent recognition, information real-time transmission and report generation, and in a specific routing inspection task, first-line routing inspection personnel only need to issue the routing inspection task to the unmanned aerial vehicle, the unmanned aerial vehicle autonomously completes the routing inspection task, and manual intervention is not needed in the whole process.
The specific structure of the invention needs to be described that the connection relation between the component modules adopted by the invention is definite and realizable, and besides the specific description in the embodiment, the specific connection relation can bring about corresponding technical effects, and on the premise of not depending on execution of corresponding software programs, the technical problems of the invention are solved, the types of the components, the modules and the specific components, the connection modes of the components and the expected technical effects brought by the technical characteristics are clear, complete and realizable, and the conventional use method and the expected technical effects brought by the technical characteristics are all disclosed in patents, journal papers, technical manuals, technical dictionaries and textbooks which can be acquired by a person in the field before the application date, or the prior art such as conventional technology, common knowledge in the field, and the like, so that the provided technical scheme is clear, complete and the corresponding entity products can be reproduced or obtained according to the technical means.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention.
Claims (9)
1. A distribution line defect intelligent identification method based on vision multi-mode fusion is characterized by comprising the following steps of: the method comprises the following steps:
s1: setting pictures acquired by the infrared camera and the zoom camera under different far and near focal lengths through the unmanned aerial vehicle, and recording focal length information of the infrared camera and the zoom camera and the corresponding acquired images;
s2: carrying out affine matrix solution on each group of acquired infrared images and RGB images, and solving affine matrixes from the RGB images to the infrared images;
s3: training all acquired RGB images by using a yolov5l target detection model, wherein two types of training are adopted, one type of training is a power transmission line component, the other type of training is a component defect, and the trained yolov5l target detection model is deployed on edge-end equipment carried by an unmanned aerial vehicle;
s4: cutting and separating infrared and visible light spliced video streams captured by an unmanned aerial vehicle, transmitting the visible light video streams to a yolov5l target detection model trained in edge equipment for detection, finally judging the detected target, directly marking the detected target in an image if the detected target is a component defect, mapping the detected target into an infrared image through an affine matrix if the detected target is a component target, solving the highest temperature and the lowest temperature in an infrared image marking frame, judging whether the component has a thermal defect or not, and marking the detected target in the image.
2. The intelligent recognition method for the defects of the distribution line based on the vision multi-mode fusion is characterized by comprising the following steps of: in the step S1, the two images acquired by the infrared camera and the zoom camera are images of the same size and position of the object acquired by the unmanned aerial vehicle at the same position and same height of the same component.
3. The intelligent recognition method for the defects of the distribution line based on the vision multi-mode fusion is characterized by comprising the following steps of: and in the step S3, a CBAM attention mechanism module is inserted into each CSP1_X module in the yolov5l basic network by the yolov5l target detection model, and important position information of a plurality of different scale feature maps is transmitted to the FPN+PAN network structure.
4. The intelligent identification method for the defects of the distribution line based on the vision multi-mode fusion as claimed in claim 3, wherein the intelligent identification method is characterized by comprising the following steps: the network structure of the backbond part of the yolov5l target detection model comprises a first CBS module, a second CBS module, a first CSP1_3 module, a first CBAM module, a third CBS module, a CSP1_6 module, a second CBAM module, a fourth CBS module, a CSP1_9 module, a third CBAM module, a fifth CBS module, a second CSP1_3 module, a fourth CBAM module and an SPPF module which are sequentially connected.
5. The intelligent recognition method for the defects of the distribution line based on the vision multi-mode fusion is characterized by comprising the following steps of: the part defects comprise body defects and channel defects, wherein the body defects comprise joint heating, heat pins, missing nuts, insulator breakage, flashover, bird nests and floaters, and the channel defects are tree barriers.
6. Intelligent distribution line defect identification system based on vision multi-mode fusion, which is characterized in that: the intelligent recognition method for the defects of the distribution lines based on the vision multi-mode fusion is characterized by comprising an unmanned aerial vehicle and a digital intelligent inspection management and control platform, wherein an infrared camera, a zoom camera, edge equipment and an RTK positioning system are mounted on the unmanned aerial vehicle, the intelligent recognition method for the defects of the distribution lines based on the vision multi-mode fusion is deployed on the edge equipment, and an LED light source is fixed on a fuselage of the unmanned aerial vehicle.
7. The intelligent recognition system for the defects of the distribution line based on the vision multi-mode fusion as claimed in claim 6, wherein: the intelligent inspection management and control platform integrates an existing line library, a 3D point cloud library of a distribution network, an avionic point library, a sample library and a model library.
8. The intelligent recognition system for the defects of the distribution line based on the vision multi-mode fusion as claimed in claim 6, wherein: the zoom camera is provided with a 25mm fixed focus lens, and the aperture is F1.8.
9. The intelligent recognition system for the defects of the distribution line based on the vision multi-mode fusion as claimed in claim 6, wherein: the LED light source is fixed on the unmanned aerial vehicle body through a machine seat capable of being remotely controlled up and down and left and right, so that a dynamic light source is formed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310216882.4A CN116245844A (en) | 2023-03-08 | 2023-03-08 | Intelligent distribution line defect identification method and system based on vision multi-mode fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310216882.4A CN116245844A (en) | 2023-03-08 | 2023-03-08 | Intelligent distribution line defect identification method and system based on vision multi-mode fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116245844A true CN116245844A (en) | 2023-06-09 |
Family
ID=86627523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310216882.4A Pending CN116245844A (en) | 2023-03-08 | 2023-03-08 | Intelligent distribution line defect identification method and system based on vision multi-mode fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116245844A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116883391A (en) * | 2023-09-05 | 2023-10-13 | 中国科学技术大学 | Two-stage distribution line defect detection method based on multi-scale sliding window |
| CN117670882A (en) * | 2024-01-31 | 2024-03-08 | 国网江西省电力有限公司电力科学研究院 | Unmanned aerial vehicle infrared automatic focusing method and system for porcelain insulator string |
-
2023
- 2023-03-08 CN CN202310216882.4A patent/CN116245844A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116883391A (en) * | 2023-09-05 | 2023-10-13 | 中国科学技术大学 | Two-stage distribution line defect detection method based on multi-scale sliding window |
| CN116883391B (en) * | 2023-09-05 | 2023-12-19 | 中国科学技术大学 | Two-stage distribution line defect detection method based on multi-scale sliding window |
| CN117670882A (en) * | 2024-01-31 | 2024-03-08 | 国网江西省电力有限公司电力科学研究院 | Unmanned aerial vehicle infrared automatic focusing method and system for porcelain insulator string |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116245844A (en) | Intelligent distribution line defect identification method and system based on vision multi-mode fusion | |
| US11017228B2 (en) | Method and arrangement for condition monitoring of an installation with operating means | |
| CN109334543B (en) | Power line inspection system and method with cooperation of power inspection vehicle and unmanned aerial vehicle | |
| CN110033453A (en) | Based on the power transmission and transformation line insulator Aerial Images fault detection method for improving YOLOv3 | |
| CN114792319B (en) | Transformer substation inspection method and system based on transformer substation image | |
| CN110580717A (en) | Unmanned aerial vehicle autonomous inspection route generation method for electric power tower | |
| CN110888457A (en) | System and method for carrying out three-dimensional inspection on power transformation equipment by using unmanned aerial vehicle and robot | |
| CN112327906A (en) | Intelligent automatic inspection system based on unmanned aerial vehicle | |
| US20120140041A1 (en) | Method and system for the remote inspection of a structure | |
| CN105139350A (en) | Ground real-time reconstruction processing system for unmanned aerial vehicle reconnaissance images | |
| CN105743017A (en) | Unmanned aerial vehicle inspection ground measurement and control station and transportation safeguard system used for power transmission line | |
| CN111952883B (en) | Power transmission line fault recognition system and method based on three-dimensional laser radar | |
| CN113589837A (en) | Electric power real-time inspection method based on edge cloud | |
| CN111244822B (en) | Fixed-wing unmanned aerial vehicle line patrol method, system and device in complex geographic environment | |
| CN108036856A (en) | Multi-rotor unmanned aerial vehicle airborne imaging spectrum instrument real-time calibration system | |
| CN114944816A (en) | Intelligent inspection system applied to photovoltaic module | |
| CN114200958A (en) | Automatic inspection system and method for photovoltaic power generation equipment | |
| CN110703809A (en) | Unmanned aerial vehicle subway tunnel fixed-point inspection method based on wireless sensor network | |
| CN212515006U (en) | Accurate positioner of unusual target suitable for unmanned aerial vehicle photovoltaic patrols and examines scene | |
| CN113269744A (en) | Infrared intelligent identification method and system applied to defects of power transmission line equipment | |
| CN117389294A (en) | Partial discharge detection system and method for airborne solar blind ultraviolet camera | |
| CN114167245B (en) | Intelligent detection method for partial discharge on surface of power transmission and transformation equipment and unmanned aerial vehicle fusion ultraviolet system | |
| CN114721432A (en) | Unmanned aerial vehicle intelligent inspection equipment based on big data and inspection image defect analysis method | |
| CN114549432A (en) | Intelligent unmanned aerial vehicle image identification method | |
| CN114326783A (en) | Unmanned aerial vehicle inspection line planning method for distributed photovoltaic power station |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |