CN116030366A - Power line inspection detection method and system - Google Patents
Power line inspection detection method and system Download PDFInfo
- Publication number
- CN116030366A CN116030366A CN202310138720.3A CN202310138720A CN116030366A CN 116030366 A CN116030366 A CN 116030366A CN 202310138720 A CN202310138720 A CN 202310138720A CN 116030366 A CN116030366 A CN 116030366A
- Authority
- CN
- China
- Prior art keywords
- inspection
- unmanned aerial
- aerial vehicle
- image
- power line
- 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
- 238000007689 inspection Methods 0.000 title claims abstract description 109
- 238000001514 detection method Methods 0.000 title claims abstract description 54
- 230000006870 function Effects 0.000 claims description 30
- 230000007547 defect Effects 0.000 claims description 23
- 238000000034 method Methods 0.000 claims description 12
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 5
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 2
- 235000008434 ginseng Nutrition 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims 4
- 239000012212 insulator Substances 0.000 description 34
- 230000006872 improvement Effects 0.000 description 4
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009960 carding Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Images
Classifications
-
- 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
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention belongs to the technical field of electric power image detection, and provides an electric power line inspection detection method and system.
Description
Technical Field
The invention belongs to the technical field of power image detection, and particularly relates to a power line inspection detection method and system.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
In recent years, the detection of power lines has been gradually advanced in an intelligent direction, and in the current detection of power lines, images are collected by unmanned aerial vehicles and identified and detected. The power line has the characteristics of wide distribution area, long transmission distance, complex and changeable topography conditions and the like, and the problem of low accuracy in detection and identification of images shot by the unmanned aerial vehicle is caused by the fact that the quality of images shot by target components in a target area is not ideal due to the fact that factors such as shooting target areas, angles and directions of the unmanned aerial vehicle change in the shooting process. In addition, the existing method for detecting the inspection target in the inspection of the power line based on deep learning, such as a YOLOv algorithm, has the problem that the extraction precision is insufficient when the features are extracted by the existing YOLOv algorithm, so that the final detection result is inaccurate.
Disclosure of Invention
In order to solve the problems, the invention provides a power line inspection and detection method and system, which acquire positioning information by acquiring three-dimensional point cloud data of a shooting point, further adjust the parameters of an unmanned aerial vehicle camera at the shooting point, and re-acquire an inspection image by a camera with the adjusted parameters, thereby improving the quality of the acquired image, and in addition, improving the detection precision by taking an improved YOLOv3 network as a defect detection network.
In order to achieve the above object, a first aspect of the present invention provides a power line inspection method, which adopts the following technical scheme:
step 1: acquiring corresponding three-dimensional point cloud data of an unmanned aerial vehicle at a shooting point to obtain corresponding positioning information, and acquiring a first inspection image shot by a camera of the unmanned aerial vehicle at the shooting point;
step 2: determining the actual distance between the inspection target and the focal length of the unmanned aerial vehicle camera in the first inspection image by using the obtained positioning information, adjusting the parameters of the camera based on the obtained actual distance, and re-acquiring a second inspection image under the shooting point based on the unmanned aerial vehicle camera after parameter adjustment;
step 3: and inputting the acquired second inspection image into a trained defect detection model, and outputting a defect detection result.
A second aspect of the present invention provides a power line inspection and detection system, comprising:
the acquisition module is used for: acquiring corresponding three-dimensional point cloud data of an unmanned aerial vehicle at a shooting point to obtain corresponding positioning information, and acquiring a first inspection image shot by a camera of the unmanned aerial vehicle at the shooting point;
parameter adjustment module: determining the actual distance between the inspection target and the focal length of the unmanned aerial vehicle camera in the first inspection image by using the obtained positioning information, adjusting the parameters of the camera based on the obtained actual distance, and re-acquiring a second inspection image under the shooting point based on the unmanned aerial vehicle camera after parameter adjustment;
and a defect detection module: and inputting the acquired second inspection image into a trained defect detection model, and outputting a defect detection result.
A third aspect of the present invention provides a computer apparatus comprising: the system comprises a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, the processor and the memory communicate through the bus when the computer device is running, and the machine-readable instructions when executed by the processor perform a power line inspection method as described in the first aspect.
In a fourth aspect, an embodiment of the present invention provides a computer readable storage medium having a computer program stored thereon, which when executed by a processor performs a power line inspection detection method as described in the first aspect above.
The beneficial effects are that:
according to the invention, the positioning information is obtained by obtaining the three-dimensional point cloud data of the shooting point, so that the parameters of the unmanned aerial vehicle camera at the shooting point are adjusted, and the inspection image is re-obtained by the camera with the adjusted parameters, so that the quality of the obtained image is improved.
In the invention, the original YOLOv3 partial sampling layer and original transmission layer are replaced by DenseNet, so that the feature transmission is enhanced, the feature fusion is promoted, the loss of the feature information of the input image in the forward transmission process of the network is effectively avoided, and the detection capability of the inspection target is improved and the detection precision is improved by adding a feature extraction layer.
According to the invention, the FocalLoss function and the balanced cross entropy function are utilized to improve the loss function of the YOLOv3 network, so that the contribution degree of class samples occupying a small number in the sample set in the training of the YOLOv3 network model can be improved, and the detection precision of the model is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
Fig. 1 is a flowchart of a power line inspection and detection method according to the present invention.
Detailed Description
The invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
Example 1
As shown in fig. 1, the present embodiment provides a power line inspection and detection method, including:
step 1: acquiring corresponding three-dimensional point cloud data of an unmanned aerial vehicle at a shooting point to obtain corresponding positioning information, and acquiring a first inspection image shot by a camera of the unmanned aerial vehicle at the shooting point;
step 2: determining the actual distance between the inspection target and the focal length of the unmanned aerial vehicle camera in the first inspection image by using the obtained positioning information, adjusting the parameters of the camera based on the obtained actual distance, and re-acquiring a second inspection image of the inspection target under the shooting point based on the unmanned aerial vehicle camera after parameter adjustment;
step 3: and inputting the acquired second inspection image into a trained defect detection model, and outputting a defect detection result.
In step 1 and step 2 of this embodiment, taking an inspection target as an insulator string as an example, the inspection route map for the power line inspection is generated by pre-carding the shooting points on the power line to be inspected and the insulator string under the shooting points, and using the existing route planning software.
And scanning the shooting points on the power line by using radar equipment on the unmanned aerial vehicle to obtain corresponding three-dimensional point cloud data, and positioning the shooting points by using the point cloud data to enable the obtained three-dimensional point cloud data to have corresponding positioning information.
In step 2 of this embodiment, a first inspection image under a shooting point is shot by using a camera on the unmanned aerial vehicle, an inspection target, i.e., an insulator string, in the first inspection image is determined by a preset target recognition model, and a target rectangular area in which the inspection target, i.e., the insulator string, is located is output. Specifically, the object recognition model may employ an existing yolov5 object detection algorithm.
Determining an actual distance between a patrol target, namely an insulator chain, in the first patrol image and a focal length of the unmanned aerial vehicle camera by using the obtained positioning information, and adjusting parameters of the camera based on the obtained actual distance, wherein the parameters are specifically as follows:
determining a focal length center of the unmanned aerial vehicle camera and an image position coordinate corresponding to an inspection target, namely an insulator string, in a first inspection image;
determining world position coordinates corresponding to position coordinates of an inspection target, namely an insulator string, in the first inspection image according to the acquired positioning information;
obtaining an actual distance between an inspection target, namely an insulator string, in a first inspection image and the focal length center of the unmanned aerial vehicle camera according to the focal length center of the camera and the obtained world position coordinates;
and adjusting unmanned aerial vehicle parameters according to the obtained actual distance between the inspection target, namely the insulator string, in the first inspection image and the focal length center of the unmanned aerial vehicle camera, so that the focal length center of the camera of the second inspection image shot by the subsequent unmanned aerial vehicle camera is the center of the inspection target, namely the insulator string.
Specifically, the focal length center of the unmanned aerial vehicle camera is the center point of the first inspection image, and the image position coordinates corresponding to the inspection target, namely the insulator string, in the first inspection image can be represented by the center point of the target rectangular area where the image position coordinates are located, so that the focal length center of the camera and the pixel coordinates, namely the image position coordinates, of the inspection target, namely the insulator string, in the first inspection image are positioned in the first inspection image.
Based on the obtained positioning information, the world position coordinates corresponding to the position coordinates of different images in the first inspection image are included, and the camera focal length and the world position coordinates of the inspection target, namely the insulator string, in the first inspection image can be respectively determined by matching the camera focal length center with the image position coordinates corresponding to the inspection target, namely the insulator string, in the first inspection image; and further calculating to obtain the actual distance between the world position coordinate of the focal length center of the camera and the world position coordinate of the inspection target, namely the insulator string, in the first inspection image. And determining matched camera adjustment parameters according to the obtained actual distance to adjust the camera. Wherein the world position coordinates include longitude and latitude and altitude.
In step 3 of this embodiment, the present embodiment uses an improved YOLOv3 network as a defect detection model, where the improved YOLOv3 network is: the original transmission layer, the 32×32 downsampling layer and the 16×16 downsampling layer of the existing YOLOv3 network are replaced by DenseNet, and a feature extraction layer is added after the first residual block of the existing YOLOv3 network.
Specifically, the original transmission layer with lower resolution of the YOLOv3 network is replaced by DenseNet, the original transmission layer adjusts the size of an input remote second inspection image from 256×256 to 512×512, the loss of the input image characteristic information in the forward propagation process of the network is avoided, meanwhile, a 32×32 downsampling layer and a 16×16 downsampling layer in the YOLOv3 network are replaced by DenseNet, a characteristic extraction layer is newly added after the first residual block of the existing YOLOv3 network, the characteristic extraction layer extracts a first characteristic image with the size of 128×128, and 4 characteristic images are added to a second characteristic image with the size of 64×64, a third characteristic image with the size of 32×32 and a fourth characteristic image with the size of 16×16, and then the characteristic extraction capability of the network is improved by fusing the high-resolution features of the low-layer characteristics and the high-semantic information features through an FPN algorithm.
In this embodiment, the original YOLOv3 Loss function is improved by adopting the Focal Loss function and the balanced cross entropy function, and the original YOLOv3 Loss function is:
wherein ,
for center loss calculated by cross entropy function, < +.>
The size loss of the square block is calculated through a mean square error function; />
Confidence loss calculated by the cross function; />
Is the classification loss calculated by the cross entropy function.
In this embodiment, the cross entropy function in the confidence loss is replaced with a FocalLoss function, which is:
wherein ,
for Focal Loss value, ++>
、/>
All are superior ginseng and jersey>
The method is used for solving the problem that the complexity of the inspection target, namely the insulator string sample is unbalanced, and the value is 0.25 in the embodiment; />
For solving the imbalance of foreground and background in the image, the value of the embodiment is 2, wherein the insulator string to be positioned is called foreground, the other parts are called background, < >>
Is a sigmoid function.
For the classification loss function of YOLOv3, an equalization cross function is adopted for improvement, and the equalization cross entropy function is as follows:
wherein ,L2 Is a balanced cross entropy value; y is a real label of an inspection target, namely an insulator string sample; q is a predicted value of an inspection target, namely an insulator chain; beta is a modulation variable for solving the problem of insulator string sample class imbalance.
In the present embodiment, the loss function used is still composed of 4 parts for calculating the predicted loss of the center point
For calculating the block size prediction loss +.>
FocalLoss function improvement ++>
And equalization entropy improvement
。
In the embodiment, the phenomenon of unbalanced complexity of foreground and background exists in the insulator string image sample shot by the unmanned aerial vehicle, wherein the number of insulator strings in the shot image is small, the distance is short, and the background is single, and the insulator string image sample belongs to easily-distinguished samples with simple foreground; the number of insulator strings is large, the distance is long, and more sundries exist in the background, which belongs to samples difficult to distinguish. The FocalLoss function can be used for solving the problem of unbalanced foreground and background in the insulator string image. In addition, in the defect detection model training, as the problem of unbalance in number exists between the positive sample of the complete insulator string image and the negative sample of the defective insulator string image, the improved loss function can enable the contribution degree of the few samples in the samples to the loss function to be improved, so that the detection accuracy is improved.
In this embodiment, training of the defect detection model is specifically:
acquiring an insulator string image sample set, wherein the insulator string image sample set comprises a positive insulator string image sample and negative insulator string image samples of different types, and labeling images in the insulator string image sample set; wherein the defect types of the insulator string include cracks, breakage, and the like.
Training the constructed defect detection model by utilizing the insulator string image sample set to obtain a trained defect detection model.
Example two
The embodiment provides a power line inspection detecting system, including:
the acquisition module is used for: acquiring corresponding three-dimensional point cloud data of an unmanned aerial vehicle at a shooting point to obtain corresponding positioning information, and acquiring a first inspection image shot by a camera of the unmanned aerial vehicle at the shooting point;
parameter adjustment module: determining the actual distance between the inspection target and the focal length of the unmanned aerial vehicle camera in the first inspection image by using the obtained positioning information, adjusting the parameters of the camera based on the obtained actual distance, and re-acquiring a second inspection image under the shooting point based on the unmanned aerial vehicle camera after parameter adjustment;
and a defect detection module: and inputting the acquired second inspection image into a trained defect detection model, and outputting a defect detection result.
Example III
The embodiment of the invention also provides computer equipment, which comprises a processor, a memory and a bus. The memory stores machine-readable instructions executable by the processor, when the computer device runs, the processor communicates with the memory through a bus, and when the machine-readable instructions are executed by the processor, a power line inspection detection method can be executed, and specific implementation manners can be referred to the method embodiments and are not described herein.
Example IV
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with a computer program, and the computer program is executed by a processor to execute the power line inspection detection method in the embodiment of the method.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random access Memory (Random AccessMemory, RAM), or the like.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The power line inspection detection method is characterized by comprising the following steps of:
step 1: acquiring corresponding three-dimensional point cloud data of an unmanned aerial vehicle at a shooting point to obtain corresponding positioning information, and acquiring a first inspection image shot by a camera of the unmanned aerial vehicle at the shooting point;
step 2: determining the actual distance between the inspection target and the focal length of the unmanned aerial vehicle camera in the first inspection image by using the obtained positioning information, adjusting the parameters of the camera based on the obtained actual distance, and re-acquiring a second inspection image under the shooting point based on the unmanned aerial vehicle camera after parameter adjustment;
step 3: and inputting the acquired second inspection image into a trained defect detection model, and outputting a defect detection result.
2. The method for detecting the inspection of the power line according to claim 1, wherein in the step 2, the actual distance between the inspection target in the first inspection image and the focal length of the unmanned aerial vehicle camera is determined by using the obtained positioning information, specifically:
determining a focal length center of the unmanned aerial vehicle camera and an image position coordinate corresponding to a patrol target in a first patrol image in the first patrol image;
determining world position coordinates corresponding to the position coordinates of the inspection target image in the first inspection image according to the acquired positioning information;
obtaining an actual distance between a patrol target in a first patrol image and the focal length center of the unmanned aerial vehicle camera according to the focal length center of the camera and the obtained world position coordinates;
and adjusting unmanned aerial vehicle parameters according to the obtained actual distance between the inspection target in the first inspection image and the focal length center of the unmanned aerial vehicle camera, so that the focal length center of the camera of the second inspection image shot by the follow-up unmanned aerial vehicle camera is the center of the inspection target.
3. The power line inspection detection method according to claim 2, wherein the focal length center of the unmanned aerial vehicle camera is the center point of the first inspection image, and the image position coordinates of the inspection target corresponding to the inspection target in the first inspection image are represented by the center point of the target rectangular area where the inspection target is located.
4. The power line inspection and detection method as claimed in claim 1, wherein the defect detection model structure is: and replacing an original transmission layer, a 32 multiplied by 32 downsampling layer and a 16 multiplied by 16 downsampling layer of the Yolov3 network with DenseNet by taking the Yolov3 network as a backbone network, and adding a feature extraction layer after the first residual block of the Yolov3 network.
5. The power line inspection and detection method according to claim 4, wherein the loss function in the YOLOv3 defect detection model is: the sum of the center Loss of the cross entropy function calculation, the size Loss of the square block of the variance function calculation, the confidence Loss of the Focal Loss function calculation, and the classification Loss of the balanced cross function calculation.
6. The power line inspection and detection method as claimed in claim 5, wherein the FocalLoss function is:
wherein ,
for Focal Loss value, ++>
、/>
All are superior ginseng and jersey>
Is a sigmoid function.
7. The power line inspection and detection method of claim 1, wherein the balanced cross entropy function is:
wherein ,L2 Is a balanced cross entropy value; y is a real label of the inspection target sample; q is a predicted value of the inspection target; beta is the modulation variable.
8. A power line inspection and detection system, comprising:
the acquisition module is used for: acquiring corresponding three-dimensional point cloud data of an unmanned aerial vehicle at a shooting point to obtain corresponding positioning information, and acquiring a first inspection image shot by a camera of the unmanned aerial vehicle at the shooting point;
parameter adjustment module: determining the actual distance between the inspection target and the focal length of the unmanned aerial vehicle camera in the first inspection image by using the obtained positioning information, adjusting the parameters of the camera based on the obtained actual distance, and re-acquiring a second inspection image under the shooting point based on the unmanned aerial vehicle camera after parameter adjustment;
and a detection module: and inputting the acquired second inspection image into a trained defect detection model, and outputting a defect detection result.
9. A computer device, comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory in communication via the bus when the computer device is running, the machine-readable instructions when executed by the processor performing a power line inspection method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, performs a power line inspection detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310138720.3A CN116030366A (en) | 2023-02-21 | 2023-02-21 | Power line inspection detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310138720.3A CN116030366A (en) | 2023-02-21 | 2023-02-21 | Power line inspection detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116030366A true CN116030366A (en) | 2023-04-28 |
Family
ID=86074099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310138720.3A Pending CN116030366A (en) | 2023-02-21 | 2023-02-21 | Power line inspection detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116030366A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633629A (en) * | 2019-08-02 | 2019-12-31 | 广东电网有限责任公司清远供电局 | Power grid inspection method, device, equipment and storage medium based on artificial intelligence |
| CN112183667A (en) * | 2020-10-31 | 2021-01-05 | 哈尔滨理工大学 | Insulator fault detection method in cooperation with deep learning |
| CN113191237A (en) * | 2021-04-21 | 2021-07-30 | 深圳职业技术学院 | Improved YOLOv 3-based fruit tree image small target detection method and device |
| CN114020039A (en) * | 2021-12-09 | 2022-02-08 | 北京中电普华信息技术有限公司 | Automatic focusing system and method for unmanned aerial vehicle inspection tower |
-
2023
- 2023-02-21 CN CN202310138720.3A patent/CN116030366A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633629A (en) * | 2019-08-02 | 2019-12-31 | 广东电网有限责任公司清远供电局 | Power grid inspection method, device, equipment and storage medium based on artificial intelligence |
| CN112183667A (en) * | 2020-10-31 | 2021-01-05 | 哈尔滨理工大学 | Insulator fault detection method in cooperation with deep learning |
| CN113191237A (en) * | 2021-04-21 | 2021-07-30 | 深圳职业技术学院 | Improved YOLOv 3-based fruit tree image small target detection method and device |
| CN114020039A (en) * | 2021-12-09 | 2022-02-08 | 北京中电普华信息技术有限公司 | Automatic focusing system and method for unmanned aerial vehicle inspection tower |
Non-Patent Citations (2)
| Title |
|---|
| 王博: "基于深度视觉的无人机自主巡检系统研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, vol. 2021, no. 2, pages 031 - 2 * |
| 颜宏文 等: "基于改进YOLOv3的绝缘子串定位与状态识别方法", 《高电压技术》, vol. 46, no. 2, pages 423 - 427 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111353413A (en) | Low-missing-report-rate defect identification method for power transmission equipment | |
| CN112767391A (en) | Power grid line part defect positioning method fusing three-dimensional point cloud and two-dimensional image | |
| CN109815300B (en) | Vehicle positioning method | |
| CN110866494B (en) | Urban group extraction method and system based on optical remote sensing image | |
| CN110910440B (en) | Power transmission line length determination method and system based on power image data | |
| CN115861853A (en) | Transmission line bird nest detection method in complex environment based on improved yolox algorithm | |
| CN110705397A (en) | YOLOv3 pruning identification insulator defect method suitable for small field sample amount | |
| Sampath et al. | Estimation of rooftop solar energy generation using Satellite Image Segmentation | |
| CN115240089A (en) | Vehicle detection method of aerial remote sensing image | |
| CN115761513A (en) | Intelligent remote sensing identification method for mountain large landslide based on semi-supervised deep learning | |
| CN117152513A (en) | Vehicle boundary positioning method for night scene | |
| CN115409789A (en) | Power transmission line engineering defect detection method based on image semantic segmentation | |
| CN113971666A (en) | Power transmission line machine inspection image self-adaptive identification method based on depth target detection | |
| CN113379719A (en) | Road defect detection method, road defect detection device, electronic equipment and storage medium | |
| CN112926415A (en) | Pedestrian avoiding system and pedestrian monitoring method | |
| CN112509321A (en) | Unmanned aerial vehicle-based driving control method and system for urban complex traffic situation and readable storage medium | |
| CN116893685A (en) | Unmanned aerial vehicle route planning method and system | |
| CN115995075A (en) | Vehicle self-adaptive navigation method and device, electronic equipment and storage medium | |
| CN116030366A (en) | Power line inspection detection method and system | |
| CN116071721A (en) | Transformer-based high-precision map real-time prediction method and system | |
| CN115937492A (en) | Transformer equipment infrared image identification method based on feature identification | |
| CN115690610A (en) | Unmanned aerial vehicle navigation method based on image matching | |
| CN115527128A (en) | Semantic segmentation based aerial image rapid positioning method | |
| JP7347696B2 (en) | Training device, control method, and program | |
| CN115015904A (en) | Speed measurement method, system and equipment based on fusion of radar and camera and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230428 |