CN117292277A - Insulator fault detection method based on binocular unmanned aerial vehicle system and deep learning - Google Patents
Insulator fault detection method based on binocular unmanned aerial vehicle system and deep learning Download PDFInfo
- Publication number
- CN117292277A CN117292277A CN202311224483.9A CN202311224483A CN117292277A CN 117292277 A CN117292277 A CN 117292277A CN 202311224483 A CN202311224483 A CN 202311224483A CN 117292277 A CN117292277 A CN 117292277A
- Authority
- CN
- China
- Prior art keywords
- insulator
- aerial vehicle
- unmanned aerial
- binocular
- vehicle system
- 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
- 239000012212 insulator Substances 0.000 title claims abstract description 77
- 238000001514 detection method Methods 0.000 title claims abstract description 56
- 238000013135 deep learning Methods 0.000 title claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 18
- 230000007246 mechanism Effects 0.000 claims abstract description 14
- 238000000034 method Methods 0.000 claims description 24
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 8
- 238000002372 labelling Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims 1
- 238000012545 processing Methods 0.000 description 13
- 238000007689 inspection Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- KLDZYURQCUYZBL-UHFFFAOYSA-N 2-[3-[(2-hydroxyphenyl)methylideneamino]propyliminomethyl]phenol Chemical compound OC1=CC=CC=C1C=NCCCN=CC1=CC=CC=C1O KLDZYURQCUYZBL-UHFFFAOYSA-N 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000004020 conductor Substances 0.000 description 1
- 239000000356 contaminant Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 201000001098 delayed sleep phase syndrome Diseases 0.000 description 1
- 208000033921 delayed sleep phase type circadian rhythm sleep disease Diseases 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000010292 electrical insulation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000035882 stress Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/17—Terrestrial scenes taken from planes or by drones
-
- 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
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Remote Sensing (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an insulator fault detection method based on a binocular unmanned aerial vehicle system and deep learning. And secondly, detecting the insulator faults through a YOLOv7 network integrating an attention mechanism, and training by using a training set. And finally, inputting the insulator fault image data to be detected into a trained model for detection, obtaining a fault detection result, and transmitting the fault type and position information to an operator or recording the fault type and position information in a patrol report after the fault is detected. The invention reduces the influence of human subjective factors on the detection result, improves the accuracy and the robustness of fault detection, and has stronger adaptability and flexibility.
Description
Technical Field
The invention relates to the technical field of insulator fault detection, in particular to an insulator fault detection method based on a binocular unmanned aerial vehicle system and deep learning.
Background
In an electrical power transmission system, insulators are critical elements used to support and secure conductors to isolate electrical insulation between a high voltage power line and a support structure. However, due to prolonged exposure to harsh environments, insulators may be subject to climatic conditions, contaminants, mechanical stress, and aging, resulting in failure or damage to the insulator surface.
In order to ensure safe and reliable operation of the power system, it is important to detect and diagnose insulator faults in time. The traditional insulator fault detection method generally depends on manual inspection, and the method has the problems of high manpower resource consumption, low working efficiency, subjective detection result and the like. Therefore, the development of an automatic, efficient and accurate insulator fault detection method has important significance.
At present, although some insulator detection methods based on unmanned aerial vehicle systems have been proposed, conventional methods often rely on monocular vision alone or use a rule-based image processing algorithm for fault detection. These methods may be affected by factors such as illumination conditions, viewing angle variation, and complex background, resulting in insufficient accuracy and robustness of fault detection.
Disclosure of Invention
The invention aims to provide an insulator fault detection method based on a binocular unmanned aerial vehicle system and deep learning, and aims to provide an automatic, efficient and accurate insulator fault detection technology. The binocular vision system carried by the unmanned aerial vehicle is utilized to collect the three-dimensional image data of the insulator, and the quick detection and classification of the surface faults of the insulator are realized through a deep learning algorithm.
The first aspect of the invention provides an insulator fault detection method based on a binocular unmanned aerial vehicle system and deep learning, which comprises the following steps:
step one: and a binocular vision unmanned aerial vehicle system is built through a binocular unmanned aerial vehicle.
Step two: and acquiring the data of the insulator images by using the binocular unmanned aerial vehicle, constructing a training set and a testing set, and preprocessing the data of the insulator images.
Step three: insulator fault detection is performed through a YOLOv7 network integrating an attention mechanism, and training is performed by using a training set.
Step four: and inputting the insulator fault image data to be detected into a trained model for detection, and obtaining a fault detection result.
Step five: upon detection of a fault, the fault type and location information may be transmitted to an operator or recorded in a patrol report.
Preferably, in the binocular vision unmanned aerial vehicle system described in the first step, two high-definition cameras are mounted on a four-wing unmanned aerial vehicle. The method comprises the steps of acquiring the data of the electronic images through two high-definition cameras, processing the data, uniformly uploading the data to a cloud storage service for storage so as to facilitate subsequent processing, and simultaneously transmitting the data of the video stream back to a binocular unmanned aerial vehicle flight controller for displaying real-time pictures.
Preferably, in the second step, fault labeling is performed on the collected data of the insulator images, and fault type and position information existing in each insulator image are labeled. After labeling is completed, a dataset is constructed that contains a large number of labeled samples. And the image is subjected to preprocessing such as denoising and color calibration so as to ensure the quality and consistency of the data.
Preferably, the third step adopts an improved bipartite K-means clustering algorithm to design the size of an anchor frame for a damaged insulator image dataset, and uses a YOLOv7 model fused with an attention mechanism to detect the insulator faults. The improved YOLOv7 model optimizes the network structure by introducing a mixed attention mechanism module ACmix, by adding a mixed attention mechanism module ACmix before the CBS module and SPPCSPC module in the YOLOv7 model. By inputting binocular vision data, faults in the insulator images are detected and classified rapidly and accurately. The weight and parameters of the model are updated through repeated iterative training processes, so that the model can be better adapted to the detection task of the insulator faults.
Preferably, in the step four, insulator fault detection and identification are carried out, and when the binocular unmanned aerial vehicle executes an insulator inspection task, two high-definition cameras mounted on the binocular unmanned aerial vehicle acquire insulator images in real time. The acquired images are input into the trained model in the third step, and faults in the insulator images are automatically detected through reasoning and recognition of the model.
A second aspect of the present invention provides an insulator fault detection device based on a binocular unmanned aerial vehicle system and deep learning, comprising: the system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the insulator fault detection method based on the binocular unmanned aerial vehicle system and deep learning when executing the program.
A third aspect of the present invention provides a computer-readable storage medium storing a computer program for executing the above-described binocular unmanned aerial vehicle system and deep learning-based insulator fault detection method.
The invention has the following benefits:
1. and (3) automatic detection: according to the method, the automatic detection and classification of the faults of the insulators are realized through a binocular vision system and a deep learning model which are carried by the unmanned aerial vehicle. Compared with the traditional manual inspection method, the invention eliminates a great deal of investment of human resources and reduces the influence of human subjective factors on the detection result.
2. High efficiency and accuracy: the YOLOv7 model, which incorporates a fused attention mechanism, enables the rapid and accurate detection of faults in the insulator image. By means of the stereoscopic image data provided by the binocular vision system, the method can obtain more depth information, and accuracy and robustness of fault detection are improved.
3. The cost is saved: compared with the traditional manual inspection method, the invention saves a great deal of manpower resources and time cost. The unmanned aerial vehicle system reduces the requirement for manual inspection, improves the detection efficiency, and reduces the maintenance and repair cost.
4. Application flexibility: the invention is suitable for various insulator inspection scenes, and can implement efficient insulator fault detection in cities or rural areas, mountain areas or other hard-to-reach areas. Meanwhile, the method can adapt to different illumination conditions, visual angle changes, complex background and other practical application environments, and has strong adaptability and flexibility.
Drawings
FIG. 1 is a flow chart of the detection process of the present invention;
FIG. 2 is a flow chart of binocular unmanned aerial vehicle image processing;
FIG. 3 is a block diagram of ACmix;
fig. 4 is a diagram of a modified YOLOv7 network structure.
Detailed Description
The invention will now be described in further detail with reference to the drawings and examples.
The embodiment of the application discloses an insulator fault detection method based on a binocular vision unmanned aerial vehicle system and deep learning, wherein a specific flow is shown in a figure 1, and the method comprises the following steps: step one: and constructing a four-wing binocular unmanned aerial vehicle system. The four-wing unmanned aerial vehicle is adopted to carry double cameras and is used for effectively collecting image data and real-time video. After the image data acquisition, the image processing flow in fig. 2 is developed. First, a Zhang Zhengyou calibration method is used for calibrating the left and right images so as to accurately determine the internal and external parameters of each camera. Then, the image is corrected to improve the accuracy of image matching and distance calculation. Then, parallax computation and matching are performed, and a corresponding point of each pixel in the two view images is found. And finally, calculating the depth value and the distance value of each pixel point based on the parallax information, and constructing the insulator image data with depth perception and high robustness. In order to facilitate the processing of the subsequent modules, the image data is integrally uploaded to a cloud storage service. Simultaneously, the real-time video stream is transmitted back to the flight controller by means of wireless communication to display the pictures in real time.
Step two: and constructing a training set and a testing set by using the insulator data acquired by the binocular unmanned aerial vehicle, and preprocessing the data. And labeling objects of different classifications by using a professional data labeling tool labelImg, and arranging the labels into a PASCAL VOC format. In the corresponding XML file, the damaged insulator is labeled "defect" and the normal insulator is labeled "insulator". In the whole network training process, the data set is distributed by randomly extracting pictures, so that the non-uniformity of data sample distribution can be ensured not to influence the accuracy of training detection results. The four coordinate information of the real frame is mapped back to the original image, so that the position of the insulator is accurately positioned, and the target path and the label can be identified.
When the network training is carried out, in order to ensure that the data distribution balance of the verification set does not influence the training effect of the network model, the total number of the training set and the verification set is kept at a ratio of 9:1 to the number of the test sets, and meanwhile, the training set and the verification set are distributed according to the same ratio of 9:1.
In addition, in order to improve the image quality, an adaptive median filtering algorithm is used for denoising. The main steps of the algorithm include: first, an initial filter window size is selected and traversed pixel by pixel starting from the upper left corner of the image. For each pixel, a corresponding neighborhood region is determined, the median value of the pixels within the neighborhood is calculated and compared to the current pixel value. If the difference exceeds the threshold, impulse noise may be present, requiring an increase in window size to recalculate the median. And according to the judging result, selecting to reserve the original pixel value or replacing the current pixel value by using the recalculated median value, thereby completing the denoising processing. The entire image is then continued to be traversed until all pixels have completed processing. According to the method, the size of the filter is automatically adjusted according to surrounding neighborhood information so as to adapt to noise of different areas, noise interference in an image is effectively reduced, and a clearer image result is obtained.
Step three: model training was performed using YOLOv7 network, which fuses the attention mechanisms.
Firstly, the implementation adopts an improved bipartite K-means clustering algorithm to design the size of an anchor frame of a damaged insulator data set. The method comprises the steps of regarding the whole data set as a single cluster, and dividing the whole data set through a binary K-means algorithm to form 2 clusters. When the number of clusters is smaller than a preset k value, calculating an error for each cluster, and selecting a cluster having the largest error for division. This operation is repeated until the number of clusters reaches a preset value to solve the problem of the sensitivity of K-Means to the initial cluster center.
Subsequently, an improvement on the YOLOv 7-based network structure, as shown in fig. 4, has introduced a new attention mechanism ACmix, the network structure of which is shown in fig. 3. ACmix combines the advantages of self-attention and convolution to improve the extraction capability of target features of the network in important areas and non-important areas, thereby reducing the occurrence of insulator missed detection. This attention mechanism is embedded into the CBS module in the Backbone and the end of the Backbone to enhance the extraction of different regional target features.
The specific flow is as follows: first, the input feature map is projected and segmented, then connected in the depth dimension, forming a rich set of intermediate features. In terms of convolution paths, channel layer full-connection is performed on the intermediate feature set to perform channel expansion, then displacement operation is performed, and finally the intermediate feature set is aggregated into feature output of corresponding dimension. The aim of the path is to effectively extract local feature information in the remote sensing image, fuse the features from different levels and finally generate the feature output in HWC format.
On the other hand, for the self-attention path, the intermediate features are clustered into N groups, each group containing three feature maps. These feature maps are generated by different 1*1 convolutions and then the three feature maps are used as query, key and value inputs for the multi-headed self-attention module. This path can not only consider image information in a global scope, but also fully pay attention to the region where the object exists, ultimately producing a characteristic output of h×w×c. And finally, combining the outputs of the two paths through splicing concat operation to obtain a required output result.
Step four: inputting the image data of the insulator faults to be detected into a trained model for detection, and obtaining the positioning of the insulator faults, wherein the steps comprise: 1. loading a model and image data: ensuring that the image data has been pre-processed as per the input requirements of YOLOv7, the image is scaled to the size required by the model. 2. And (3) predicting an operation model: and predicting the image data to be detected by using the loaded model. And returning the position and the confidence score of the detected target boundary box positioning fault. 3. Interpretation results: and according to the output of the model prediction, explaining whether the image has faults and the positions and types of the faults.
And finally, outputting the detection result, drawing a marked result image on the image, storing a data file of the detection result, generating a patrol report and sending an maintainer.
By adopting the insulator fault detection method based on the binocular vision unmanned aerial vehicle system and the improved YOLOv7 model, the invention can realize rapid and accurate nondestructive detection of the insulator and improve the reliability and safety of the power transmission system. Meanwhile, the method reduces the dependence on manual inspection, saves human resources, improves the efficiency of fault detection, and is widely applicable to inspection application scenes of insulators.
The embodiment of the application also discloses insulator fault detection equipment based on binocular unmanned aerial vehicle system and deep learning, include: the system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the insulator fault detection method based on the binocular unmanned aerial vehicle system and deep learning when executing the program.
The embodiment of the application also discloses a computer readable storage medium, wherein the storage medium stores a computer program, and the computer program is used for executing the insulator fault detection method based on the binocular unmanned aerial vehicle system and the deep learning.
Those skilled in the art will appreciate that implementing all or part of the above-described methods may be accomplished by way of a computer program, which may be stored on a non-transitory computer readable storage medium and which, when executed, may comprise the steps of the above-described embodiments of the methods. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
A processor in the present application may include one or more processing cores. The processor performs the various functions of the present application and processes the data by executing or executing instructions, programs, code sets, or instruction sets stored in memory, calling data stored in memory. The processor may be at least one of an application specific integrated circuit (Application Specific IntegratedCircuit, ASIC), a digital signal processor (Digital Signal Processor, DSP), a digital signal processing device (Digital Signal Processing Device, DSPD), a Programmable logic device (Programmable LogicDevice, PLD), a field Programmable gate array (Field Programmable Gate Array, FPGA), a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, and a microprocessor. It will be appreciated that the electronic device for implementing the above-mentioned processor function may be other for different apparatuses, and embodiments of the present application are not specifically limited.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-described division of the functional units and modules is illustrated, and in practical application, the above-described functional distribution may be performed by different functional units and modules according to needs, i.e. the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-described functions.
The foregoing are all preferred embodiments of the present application, and are not intended to limit the scope of the present application in any way, therefore: all equivalent changes in structure, shape and principle of this application should be covered in the protection scope of this application.
Claims (8)
1. The insulator fault detection method based on the binocular unmanned aerial vehicle system and the deep learning is characterized by comprising the following steps of:
step one: building a binocular vision unmanned aerial vehicle system through a binocular unmanned aerial vehicle;
step two: acquiring the data of the insulator image by using a binocular unmanned aerial vehicle, constructing a training set and a testing set, and preprocessing the data of the insulator image;
step three: performing insulator fault detection through a YOLOv7 network integrating an attention mechanism, and training by using a training set;
step four: inputting the insulator fault image data to be detected into a trained model for detection, and obtaining a fault detection result;
step five: after detecting the fault, the fault type and position information is transmitted to an operator or recorded in a patrol report.
2. The method for detecting the faults of the insulators based on the binocular unmanned aerial vehicle system and the deep learning according to claim 1 is characterized in that in the binocular vision unmanned aerial vehicle system in the first step, a four-wing unmanned aerial vehicle is adopted to carry two high-definition cameras, the insulator image data are obtained through the two high-definition cameras, the insulator image data are processed and then are uniformly uploaded to a cloud storage service for storage, and meanwhile video stream data are transmitted back to a binocular unmanned aerial vehicle flight controller for displaying real-time pictures.
3. The method for detecting the insulator fault based on the binocular unmanned aerial vehicle system and the deep learning according to claim 1, wherein the preprocessing in the second step comprises denoising and color calibration of the image;
and step two, performing fault labeling on the acquired insulator image data, and labeling fault type and position information existing in each insulator image.
4. The insulator fault detection method based on the binocular unmanned aerial vehicle system and the deep learning according to claim 1, wherein the third specific process is as follows:
3.1, designing the size of an anchor frame for the damaged insulator image dataset by adopting a clustering algorithm;
3.2, detecting the faults of the insulators by using a YOLOv7 model of a fusion attention mechanism, and detecting and classifying the faults in the images of the insulators;
and 3.3, updating the weight and the parameters of the YOLOv7 model fusing the attention mechanism through repeated iterative training.
5. The method for detecting an insulator fault based on a binocular unmanned aerial vehicle system and deep learning according to claim 4, wherein the clustering algorithm of 3.1 specifically operates as: firstly, the whole data set is regarded as a single cluster, and is divided by a binary K-means algorithm to form two clusters; secondly, calculating errors for each cluster when the number of the clusters is smaller than a preset k value, and selecting the cluster with the largest error for division; finally, this operation is repeated until the number of clusters reaches a preset value.
6. The method for detecting an insulator fault based on a binocular unmanned aerial vehicle system and deep learning according to claim 4, wherein the YOLOv7 model of the fused attention mechanism of 3.2 is specifically: by adding a mixed attention mechanism module ACmix in the CBS module of the YOLOv7 model and before the SPPCSPC module.
7. Insulator fault detection equipment based on binocular unmanned aerial vehicle system and degree of depth study, its characterized in that includes: a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the binocular unmanned aerial vehicle system and deep learning-based insulator fault detection method of any one of claims 1-6 when the processor executes the program.
8. A computer readable storage medium, characterized in that the storage medium stores a computer program for executing the binocular unmanned aerial vehicle system and deep learning-based insulator fault detection method of any one of the above claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311224483.9A CN117292277A (en) | 2023-09-21 | 2023-09-21 | Insulator fault detection method based on binocular unmanned aerial vehicle system and deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311224483.9A CN117292277A (en) | 2023-09-21 | 2023-09-21 | Insulator fault detection method based on binocular unmanned aerial vehicle system and deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117292277A true CN117292277A (en) | 2023-12-26 |
Family
ID=89240239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311224483.9A Pending CN117292277A (en) | 2023-09-21 | 2023-09-21 | Insulator fault detection method based on binocular unmanned aerial vehicle system and deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117292277A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117726990A (en) * | 2023-12-27 | 2024-03-19 | 浙江恒逸石化有限公司 | Method and device for detecting spinning workshop, electronic equipment and storage medium |
-
2023
- 2023-09-21 CN CN202311224483.9A patent/CN117292277A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117726990A (en) * | 2023-12-27 | 2024-03-19 | 浙江恒逸石化有限公司 | Method and device for detecting spinning workshop, electronic equipment and storage medium |
| CN117726990B (en) * | 2023-12-27 | 2024-05-03 | 浙江恒逸石化有限公司 | Method and device for detecting spinning workshop, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111507958B (en) | Target detection method, training method of detection model and electronic equipment | |
| CN111798407B (en) | Electrified equipment fault diagnosis method based on neural network model | |
| CN111797890A (en) | Method and system for detecting defects of power transmission line equipment | |
| CN111680746B (en) | Vehicle damage detection model training, vehicle damage detection method, device, equipment and medium | |
| CN111160301B (en) | Tunnel disease target intelligent identification and extraction method based on machine vision | |
| CN112183788A (en) | Domain adaptive equipment operation detection system and method | |
| US20200394415A1 (en) | Object detection apparatus, prediction model generation apparatus, object detection method, and program | |
| CN110853005A (en) | Immunohistochemical membrane staining section diagnosis method and device | |
| CN117292277A (en) | Insulator fault detection method based on binocular unmanned aerial vehicle system and deep learning | |
| CN111524063A (en) | Remote sensing image fusion method and device | |
| CN114494185B (en) | Electrical equipment fault detection method based on RGB-T multi-scale feature fusion | |
| CN112766110A (en) | Training method of object defect recognition model, object defect recognition method and device | |
| CN110910360A (en) | Power grid image positioning method and image positioning model training method | |
| CN110807758B (en) | Method, device, equipment and storage medium for detecting uncovered area of heat preservation quilt | |
| CN111931721B (en) | Method and device for detecting color and number of annual inspection label and electronic equipment | |
| CN117456198A (en) | Power transmission line fault detection method based on improved Yolov5 neural network | |
| CN112183382A (en) | Unmanned traffic light detection and classification method and device | |
| CN112465735A (en) | Pedestrian detection method, device and computer-readable storage medium | |
| CN116934737A (en) | Weld joint combination defect identification and classification method | |
| CN114529543B (en) | Installation detection method and device for peripheral screw gasket of aero-engine | |
| CN115690505A (en) | Photovoltaic module fault detection method and device, computer equipment and storage medium | |
| CN115147591A (en) | Transformer equipment infrared image voltage heating type defect diagnosis method and system | |
| KR102210571B1 (en) | Bridge and tunnel safety diagnosis remote monitoring alarm method using GPS coordinates and mobile communication system | |
| CN115601293A (en) | Object detection method and device, electronic equipment and readable storage medium | |
| CN114998889A (en) | Intelligent identification method and system for immersive three-dimensional image |
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 |