CN113033489B - Power transmission line insulator identification positioning method based on lightweight deep learning algorithm - Google Patents
Power transmission line insulator identification positioning method based on lightweight deep learning algorithm Download PDFInfo
- Publication number
- CN113033489B CN113033489B CN202110443224.XA CN202110443224A CN113033489B CN 113033489 B CN113033489 B CN 113033489B CN 202110443224 A CN202110443224 A CN 202110443224A CN 113033489 B CN113033489 B CN 113033489B
- Authority
- CN
- China
- Prior art keywords
- transmission line
- power transmission
- insulator
- model
- line insulator
- 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.)
- Active
Links
- 239000012212 insulator Substances 0.000 title claims abstract description 90
- 230000005540 biological transmission Effects 0.000 title claims abstract description 82
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000013135 deep learning Methods 0.000 title claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000012360 testing method Methods 0.000 claims description 12
- 239000013598 vector Substances 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 9
- 230000004927 fusion Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 4
- 238000011176 pooling Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 15
- 238000007689 inspection Methods 0.000 abstract description 8
- 238000010586 diagram Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/182—Network patterns, e.g. roads or rivers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a recognition and positioning method for a power transmission line insulator based on a lightweight deep learning algorithm, which comprises the steps of collecting visible light channel pictures of the power transmission line insulator on site to construct a visible light image dataset of the power transmission line insulator; establishing an EfficientDet-d0 algorithm model, pre-training the model on an ImageNet data set, and performing fine-tuning training on the pre-trained model by utilizing a visible light image data set; and inputting the transmission line insulator image to be detected into the model after fine adjustment training, and carrying out identification and positioning on the transmission line insulator. The power transmission line insulator identification and positioning method based on the lightweight deep learning algorithm provided by the invention reduces the parameter size of the model while ensuring that the accuracy and recall rate are in the allowable range of engineering application, improves the calculation efficiency of the model, can be conveniently deployed in an embedded calculation platform, and provides technical support for unmanned aerial vehicle inspection of the power transmission line insulator.
Description
Technical Field
The invention relates to the technical field of power transmission line insulator identification and positioning, in particular to a power transmission line insulator identification and positioning method based on a lightweight deep learning algorithm.
Background
An insulator is one of the most commonly used power devices on a transmission line, and is important for maintaining the safety and stability of a power system. Positioning and identification of the insulator are the preconditions for fault diagnosis. At present, the technology of inspection modes such as unmanned aerial vehicle, robot and the like is updated, the problems of low efficiency, poor instantaneity, high risk coefficient and the like of traditional power inspection can be effectively solved, but a large amount of inspection image video data is piled up. The vast image data is extremely easy to overstock under the low-efficiency screening treatment, the real-time performance of equipment operation and maintenance is reduced, and the development of an intelligent power system is difficult to adapt.
Currently, computer vision-based deep learning algorithms are constantly alternated and revolutionized in the field of target recognition, and good results are obtained in the field of image processing. Compared with the traditional target recognition algorithm, the target recognition algorithm based on the deep convolutional neural network, such as Faster-RCNN, SSD and YOLO, automatically learns target features from a large amount of image data without manually designing a feature extractor. The end-to-end learning strategy effectively simplifies the flow of the algorithm, and improves the efficiency, accuracy and generalization capability of the target recognition algorithm. However, the deep convolutional neural network model is very complex and has too many parameters to be deployed on an embedded platform. The Efficientdet algorithm is a newer algorithm in the field of target identification, has excellent speed and precision, and is not applied in the field of insulator identification at present. Therefore, it is necessary to design a power transmission line insulator identification positioning method based on a lightweight deep learning algorithm EfficientDet with excellent speed and precision.
Disclosure of Invention
The invention aims to provide a lightweight deep learning algorithm-based power transmission line insulator identification positioning method, which reduces the parameter size of a model while ensuring that the accuracy and recall rate are within the allowable range of engineering application, improves the calculation efficiency of the model, can be conveniently deployed in an embedded calculation platform, and provides technical support for unmanned aerial vehicle inspection of the power transmission line insulator.
In order to achieve the above object, the present invention provides the following solutions:
a power transmission line insulator identification positioning method based on a lightweight deep learning algorithm comprises the following steps:
step 1: the method comprises the steps of collecting visible light channel pictures of an insulator of a power transmission line on site, and constructing a visible light image dataset of the insulator of the power transmission line;
step 2: establishing an EfficientDet-d0 algorithm model, pre-training the EfficientDet-d0 algorithm model on an ImageNet data set, and performing fine tuning training on model parameters of the pre-trained EfficientDet-d0 algorithm model by using the visible light image data set of the transmission line insulator constructed in the step 1 after the pre-training is completed;
step 3: and (3) inputting the image of the transmission line insulator to be detected into the EfficientDet-d0 algorithm model subjected to fine tuning training in the step (2) to identify and position the transmission line insulator.
Optionally, in step 1, a visible light image dataset of an insulator of the power transmission line is constructed, specifically:
collecting sample pictures of insulators in the power transmission line with 110kV, 220kV and 500kV voltage levels at different places and under different weather conditions in the power transmission line, constructing a visible light image dataset of the insulators of the power transmission line, and dividing the visible light image dataset into a training set and a testing set, wherein the training set accounts for 90% of the total number of the pictures, and the testing set accounts for 10% of the total number of the pictures.
Optionally, a visible light image dataset of the transmission line insulator is manufactured by using label software, the transmission line insulators in the visible light image dataset of the transmission line insulator are manually selected one by one, a label name insulator is input, a corresponding xml file is generated, and the model parameters of an Efficientdet-d0 algorithm model are used for fine tuning training.
Optionally, the deep learning framework of the Efficientdet-d0 algorithm model is Pytorch, the total training times are not set, and the overfitting is relieved based on an early stop method, wherein the learning rate is set to be 1E-4 between 0 and 16000 iterations, 1E-5 between 16001 and 20000 iterations, 1E-6 between 20001 and 22000, 1E-7 between 22000 and 23001, and 1E-8 between 23001 and 40000 iterations.
Optionally, in step 3, the insulator of the power transmission line is identified and positioned, specifically:
the EfficientDet-d0 algorithm model comprises a EfficientNet, biFPN pre-trained on an ImageNet, a classification and detection frame prediction network, an electric transmission line insulator image to be detected is input into the EfficientDet-d0 algorithm model EfficientNet, convolution and maximum pooling alternate processing are carried out, P3-P7 common 5 layers of feature vectors are output, the obtained P3-P7 common 5 layers of feature vectors are input into BiFPN of the EfficientDet-d0 algorithm model for multiple times to perform top-down and bottom-up feature fusion, finally 5 fusion feature vectors are output, the obtained 5 fusion feature vectors are input into the classification and detection frame prediction network respectively to obtain a prediction frame, and the insulator probability and the position of the insulator frame are obtained according to the prediction frame.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: the power transmission line insulator identification positioning method based on the lightweight deep learning algorithm provided by the invention has the advantages that the accuracy and recall rate are ensured to be within the allowable range of engineering application, the parameter size of a model is greatly reduced, the calculation efficiency of the model is improved, the model can be conveniently deployed in an embedded calculation platform, a new thought and method are provided for the inspection of the power transmission line insulator, and a certain reference value is provided for the application of edge calculation in the operation and maintenance of the power transmission line; according to the method, a visible light channel picture of the transmission line insulator is collected on site, a visible light image dataset of the transmission line insulator is constructed, the built EfficientDet-d0 algorithm model is trained through the ImageNet dataset and the visible light image dataset of the transmission line insulator, after training is finished, the transmission line insulator image to be detected is input into the EfficientDet-d0 algorithm model, identification and positioning of the transmission line insulator are carried out, identification and frame selection of the insulator can be carried out rapidly, and the EfficientDet algorithm is adopted, so that accuracy and speed are excellent, compared with other algorithms, the flow is simplified, and the efficiency, accuracy and generalization capability of target identification are improved; dividing a visible light image data set of the power transmission line insulator into a training set and a testing set, wherein the training set accounts for 90% of the total number of the photos, the testing set accounts for 10% of the total number of the photos, the training set is used for training the model, and the testing set is used for evaluating the performance of the model; the model selects a deep learning frame as Pytorch, and the overfitting is effectively relieved by an early-stop method; compared with the AP value, FLOPs and model size of the main stream YOLOv3 model, the model size recorded by the invention is about one fifteen times of the model size recorded by the invention, and the parameter size of the model is greatly reduced under the condition of ensuring the accuracy and recall rate, so that the model can be conveniently deployed in an embedded computing platform.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a power transmission line insulator identification and positioning method based on a lightweight deep learning algorithm in an embodiment of the invention;
FIG. 2 is a schematic diagram of the architecture of an Efficientdet-d0 algorithm model;
FIG. 3 is a schematic structural diagram of the MBConv module;
FIG. 4 is a schematic diagram of the output of the Efficientdet-d0 algorithm model.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a lightweight deep learning algorithm-based power transmission line insulator identification positioning method, which reduces the parameter size of a model while ensuring that the accuracy and recall rate are within the allowable range of engineering application, improves the calculation efficiency of the model, can be conveniently deployed in an embedded calculation platform, and provides technical support for unmanned aerial vehicle inspection of the power transmission line insulator.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
As shown in fig. 1-4, the method for identifying and positioning the power transmission line insulator based on the lightweight deep learning algorithm provided by the embodiment of the invention, as shown in fig. 1, comprises the following steps:
step 1: the method comprises the steps of collecting visible light channel pictures of an insulator of a power transmission line on site, and constructing a visible light image dataset of the insulator of the power transmission line;
step 2: establishing an EfficientDet-d0 algorithm model, pre-training the EfficientDet-d0 algorithm model on an ImageNet data set, and performing fine tuning training on model parameters of the pre-trained EfficientDet-d0 algorithm model by using the visible light image data set of the transmission line insulator constructed in the step 1 after the pre-training is completed;
step 3: and (3) inputting the image of the transmission line insulator to be detected into the EfficientDet-d0 algorithm model subjected to fine tuning training in the step (2) to identify and position the transmission line insulator.
In the step 1, a visible light image data set of an insulator of a power transmission line is constructed, specifically:
in order to ensure diversity of data sets, sample pictures of insulators in the transmission line with various voltage levels of 110kV, 220kV, 500kV and the like in different places and under different weather conditions are collected in the transmission line, a visible light image data set of the transmission line insulator is constructed and divided into a training set and a testing set, wherein the training set accounts for 90% of the total number of pictures, and the testing set accounts for 10% of the total number of pictures.
And (3) manufacturing a visible light image data set of the transmission line insulator by using label software, manually selecting the transmission line insulators in the visible light image data set of the transmission line insulator one by one, inputting a label name insulator, and generating a corresponding xml file for carrying out fine adjustment training on model parameters of an Efficientdet-d0 algorithm model.
The deep learning framework of the EfficientDet-d0 algorithm model is Pytorch, the model is not set with total training times, and early stop method is adopted to relieve overfitting, wherein the learning rate is set to be 1E-4 between 0 and 16000 iterations, 1E-5 between 16001 and 20000 iterations, 1E-6 between 20001 and 22000, 1E-7 between 22000 and 23001, and 1E-8 between 23001 and 40000 iterations.
In step 3, the insulator of the transmission line is identified and positioned, specifically:
as shown in fig. 2, the network structure of the afflicientdet-d 0 algorithm model is composed of three parts, wherein the first part is the afflicientnet pre-trained on the ImageNet, the first part is used as a backbone network, the second part is a BiFPN as a feature extraction network, and the third part is a classification and detection frame prediction network; inputting an electric transmission line insulator image to be detected into an EfficientDet-d0 algorithm model EfficientNet, carrying out convolution and maximum value pooling alternating processing, outputting P3-P7 total 5 layers of feature vectors, inputting the obtained P3-P7 total 5 layers of feature vectors into BiFPN of the EfficientDet-d0 algorithm model to carry out top-down and bottom-up feature fusion for multiple times, finally outputting the same 5 fused feature vectors, and inputting the obtained 5 fused feature vectors into a classification and detection frame prediction network respectively to obtain a prediction frame, wherein the insulator probability and the position of the insulator frame are obtained according to the prediction frame as shown in figure 4.
The Efficientdet algorithm uses a weighted bi-directional feature pyramid network to quickly realize multi-scale feature fusion, plays a significant role in small target detection, and most of the traditional detection algorithms adopt a top-down feature pyramid network (Feature Pyramid Networks for Object Detection, FPN), but the Efficientdet algorithm adds an extra bottom-up flow at the cost of increasing the calculation amount by replacing the FPN with PANet (Path Aggregation Network) and the like, and a new bi-directional feature network BiFPN is provided, which fuses the multi-level feature fusion concept of the FPN and the PANet, so that information can flow in top-down and bottom-up directions, and regular and efficient connection is used.
The backbone network EfficientNet consists of a mobile rollover bottleneck convolution MBConv module, as shown in fig. 3, wherein the MBConv module comprises a structure similar to a residual structure, a SE attention module is used in a short connection part, and a DropConnect method is used for randomly discarding the input of hidden nodes.
The model proposed by the invention is compared with the current mainstream 3 model to obtain the following table,
table 1 model comparison table
| Model | AP value | FLOPs | Model size |
| EfficientDet-d0 | 86.92 | 2.5B | 16MB |
| YOLOv3 | 84.23 | 71B | 235.5MB |
As shown in Table 1, the model size provided by the invention is about fifteen times of that of the YOLOv3 model, so that the parameter size of the model is greatly reduced while the accuracy and recall rate are ensured to be within the allowable range of engineering application, and the calculation efficiency of the model is improved, so that the model can be conveniently deployed in an embedded calculation platform.
The power transmission line insulator identification positioning method based on the lightweight deep learning algorithm provided by the invention has the advantages that the accuracy and recall rate are ensured to be within the allowable range of engineering application, the parameter size of a model is greatly reduced, the calculation efficiency of the model is improved, the model can be conveniently deployed in an embedded calculation platform, a new thought and method are provided for the inspection of the power transmission line insulator, and a certain reference value is provided for the application of edge calculation in the operation and maintenance of the power transmission line; according to the method, a visible light channel picture of the transmission line insulator is collected on site, a visible light image dataset of the transmission line insulator is constructed, the built EfficientDet-d0 algorithm model is trained through the ImageNet dataset and the visible light image dataset of the transmission line insulator, after training is finished, the transmission line insulator image to be detected is input into the EfficientDet-d0 algorithm model, identification and positioning of the transmission line insulator are carried out, identification and frame selection of the insulator can be carried out rapidly, and the EfficientDet algorithm is adopted, so that accuracy and speed are excellent, compared with other algorithms, the flow is simplified, and the efficiency, accuracy and generalization capability of target identification are improved; dividing a visible light image data set of the power transmission line insulator into a training set and a testing set, wherein the training set accounts for 90% of the total number of the photos, the testing set accounts for 10% of the total number of the photos, the training set is used for training the model, and the testing set is used for evaluating the performance of the model; the model selects a deep learning frame as Pytorch, and the overfitting is effectively relieved by an early-stop method; compared with the AP value, FLOPs and model size of the main stream YOLOv3 model, the model size recorded by the invention is about one fifteen times of the model size recorded by the invention, and the parameter size of the model is greatly reduced under the condition of ensuring the accuracy and recall rate, so that the model can be conveniently deployed in an embedded computing platform.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (3)
1. A power transmission line insulator identification positioning method based on a lightweight deep learning algorithm is characterized by comprising the following steps:
step 1: the method comprises the steps of collecting visible light channel pictures of an insulator of a power transmission line on site, and constructing a visible light image dataset of the insulator of the power transmission line;
step 2: establishing an EfficientDet-d0 algorithm model, pre-training the EfficientDet-d0 algorithm model on an ImageNet data set, and performing fine tuning training on model parameters of the pre-trained EfficientDet-d0 algorithm model by using the visible light image data set of the transmission line insulator constructed in the step 1 after the pre-training is completed;
the deep learning framework of the Efficientdet-d0 algorithm model is Pytorch, the total training times are not set, and the overfitting is relieved based on an early stop method, wherein the learning rate is set to be 1E-4 between 0 and 16000 iterations, 1E-5 between 16001 and 20000 iterations, 1E-6 between 20001 and 22000, 1E-7 between 22000 and 23001 and 1E-8 between 23001 and 40000 iterations;
step 3: inputting the image of the transmission line insulator to be detected into the EfficientDet-d0 algorithm model subjected to fine tuning training in the step 2, and carrying out identification and positioning on the transmission line insulator;
the EfficientDet-d0 algorithm model comprises a EfficientNet, biFPN pre-trained on an ImageNet, a classification and detection frame prediction network, an electric transmission line insulator image to be detected is input into the EfficientDet-d0 algorithm model EfficientNet, convolution and maximum pooling alternate processing are carried out, P3-P7 common 5 layers of feature vectors are output, the obtained P3-P7 common 5 layers of feature vectors are input into BiFPN of the EfficientDet-d0 algorithm model for multiple times to perform top-down and bottom-up feature fusion, finally 5 fusion feature vectors are output, the obtained 5 fusion feature vectors are input into the classification and detection frame prediction network respectively to obtain a prediction frame, and the insulator probability and the position of the insulator frame are obtained according to the prediction frame.
2. The method for identifying and positioning the transmission line insulator based on the lightweight deep learning algorithm according to claim 1, wherein in step 1, a transmission line insulator visible light image dataset is constructed, specifically:
collecting sample pictures of insulators in the power transmission line with 110kV, 220kV and 500kV voltage levels at different places and under different weather conditions in the power transmission line, constructing a visible light image dataset of the insulators of the power transmission line, and dividing the visible light image dataset into a training set and a testing set, wherein the training set accounts for 90% of the total number of the pictures, and the testing set accounts for 10% of the total number of the pictures.
3. The method for identifying and positioning the power transmission line insulator based on the lightweight deep learning algorithm according to claim 2, wherein a label software is used for manufacturing a visible light image dataset of the power transmission line insulator, the power transmission line insulators in the visible light image dataset of the power transmission line insulator are manually selected one by one in a frame mode, a label name insulator is input, a corresponding xml file is generated, and the model parameters of an Efficientdet-d0 algorithm model are used for fine tuning training.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110443224.XA CN113033489B (en) | 2021-04-23 | 2021-04-23 | Power transmission line insulator identification positioning method based on lightweight deep learning algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110443224.XA CN113033489B (en) | 2021-04-23 | 2021-04-23 | Power transmission line insulator identification positioning method based on lightweight deep learning algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113033489A CN113033489A (en) | 2021-06-25 |
| CN113033489B true CN113033489B (en) | 2024-02-13 |
Family
ID=76457540
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110443224.XA Active CN113033489B (en) | 2021-04-23 | 2021-04-23 | Power transmission line insulator identification positioning method based on lightweight deep learning algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113033489B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113780371B (en) * | 2021-08-24 | 2024-06-18 | 上海电力大学 | Insulator state edge identification method based on edge calculation and deep learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528595A (en) * | 2016-02-01 | 2016-04-27 | 成都通甲优博科技有限责任公司 | Method for identifying and positioning power transmission line insulators in unmanned aerial vehicle aerial images |
| CN110197475A (en) * | 2018-10-31 | 2019-09-03 | 国网宁夏电力有限公司检修公司 | Insulator automatic recognition system, method and application in a kind of transmission line of electricity |
| CN112541389A (en) * | 2020-09-29 | 2021-03-23 | 西安交通大学 | Power transmission line fault detection method based on EfficientDet network |
| CN112560634A (en) * | 2020-12-10 | 2021-03-26 | 齐鲁工业大学 | Method and system for rapidly detecting and positioning power line insulator based on field image |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210073692A1 (en) * | 2016-06-12 | 2021-03-11 | Green Grid Inc. | Method and system for utility infrastructure condition monitoring, detection and response |
| US10627524B2 (en) * | 2016-12-06 | 2020-04-21 | At&T Intellectual Property I, L.P. | Method and apparatus for positioning via unmanned aerial vehicles |
-
2021
- 2021-04-23 CN CN202110443224.XA patent/CN113033489B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105528595A (en) * | 2016-02-01 | 2016-04-27 | 成都通甲优博科技有限责任公司 | Method for identifying and positioning power transmission line insulators in unmanned aerial vehicle aerial images |
| CN110197475A (en) * | 2018-10-31 | 2019-09-03 | 国网宁夏电力有限公司检修公司 | Insulator automatic recognition system, method and application in a kind of transmission line of electricity |
| CN112541389A (en) * | 2020-09-29 | 2021-03-23 | 西安交通大学 | Power transmission line fault detection method based on EfficientDet network |
| CN112560634A (en) * | 2020-12-10 | 2021-03-26 | 齐鲁工业大学 | Method and system for rapidly detecting and positioning power line insulator based on field image |
Non-Patent Citations (8)
| Title |
|---|
| Hantao Tao ; .Research of Insulator Fault Identification Method Based on Atlas Intelligent Computing Platform.IEEE.2020,全文. * |
| Icing Condition Assessment of In-Service Glass Insulators Based on Graphical Shed Spacing and Graphical Shed Overhang;Yanpeng Hao;Jie Wei;Xiaolan Jiang;Lin Yang;Licheng Li;Junke Wang;Hao Li;Ruihai Li;Energies;第11卷(第2期);全文 * |
| Recognition of the Center Position of Bolt Hole in the Stand of Insulator Using Multilayer Neural Network;Kyoung Kwan Ahn;Sung Man Pyo;ournal of Institute of Control, Robotics and Systems;第9卷(第4期);全文 * |
| 付炜平 ; 施凤祥 ; 王伟 ; 刘云鹏 ; 纪欣欣.基于颜色矩阵的绝缘子单片红外图像故障诊断方法 .电瓷避雷器.2018,全文. * |
| 刘逸凡 ; 王淑青 ; 庆毅辉 ; 王晨曦 ; 兰天泽 ; 要若天.基于EfficientDet和双目摄像头的绝缘子缺陷检测.中国电力.2020,全文. * |
| 基于深度学习的复杂背景下绝缘子图像识别及定位;卢胜标;夏良标;石英;韩西坪;庞统;莫止范;刘晓伟;李德洋;田丁;中文信息(第001期);全文 * |
| 基于深度学习的输电线路绝缘子缺陷检测研究;丘灵华;朱铮涛;计算机应用研究(第0s1期);全文 * |
| 航拍图像中绝缘子的识别与故障诊断;姜浩然;金立军;闫书佳;机电工程;第32卷(第002期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113033489A (en) | 2021-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110827251B (en) | Power transmission line locking pin defect detection method based on aerial image | |
| CN110263705B (en) | Two-stage high-resolution remote sensing image change detection system oriented to remote sensing technical field | |
| CN107316054A (en) | Non-standard character recognition methods based on convolutional neural networks and SVMs | |
| CN112381097A (en) | Scene semantic segmentation method based on deep learning | |
| CN110874550A (en) | Data processing method, device, equipment and system | |
| CN112381763A (en) | Surface defect detection method | |
| CN112288700A (en) | Rail defect detection method | |
| CN115272242B (en) | YOLOv 5-based optical remote sensing image target detection method | |
| CN113033489B (en) | Power transmission line insulator identification positioning method based on lightweight deep learning algorithm | |
| CN115410087A (en) | Transmission line foreign matter detection method based on improved YOLOv4 | |
| CN114463257A (en) | Power equipment infrared image detection method and system based on deep learning | |
| CN114862768A (en) | Improved YOLOv5-LITE lightweight-based power distribution assembly defect identification method | |
| CN113837994A (en) | Photovoltaic panel defect diagnosis method based on edge detection convolutional neural network | |
| CN114819141A (en) | Intelligent pruning method and system for deep network compression | |
| CN114359167A (en) | Insulator defect detection method based on lightweight YOLOv4 in complex scene | |
| CN111027542A (en) | Target detection method improved based on fast RCNN algorithm | |
| CN114331837A (en) | Method for processing and storing panoramic monitoring image of protection system of extra-high voltage converter station | |
| CN114419005A (en) | Crack automatic detection method based on improved light weight CNN and transfer learning | |
| CN111199255A (en) | Small target detection network model and detection method based on dark net53 network | |
| CN113989296A (en) | Unmanned aerial vehicle wheat field remote sensing image segmentation method based on improved U-net network | |
| CN116485802B (en) | Insulator flashover defect detection method, device, equipment and storage medium | |
| CN113537306A (en) | Image classification method based on progressive growth element learning | |
| Wang et al. | High-Voltage Transmission Line Foreign Object and Power Component Defect Detection Based on Improved YOLOv5 | |
| CN115829029A (en) | Channel attention-based self-distillation implementation method | |
| CN115937079A (en) | YOLO v 3-based rapid detection method for defects of power transmission line |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |