CN114979184A - Method and system for online automatic inspection of electrical manual wiring quality - Google Patents
Method and system for online automatic inspection of electrical manual wiring quality Download PDFInfo
- Publication number
- CN114979184A CN114979184A CN202210442216.8A CN202210442216A CN114979184A CN 114979184 A CN114979184 A CN 114979184A CN 202210442216 A CN202210442216 A CN 202210442216A CN 114979184 A CN114979184 A CN 114979184A
- Authority
- CN
- China
- Prior art keywords
- wiring
- electrical
- digital image
- neural network
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1097—Protocols in which an application is distributed across nodes in the network for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06398—Performance of employee with respect to a job function
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- 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)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Educational Administration (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Signal Processing (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for automatically checking the quality of electric manual wiring on line, wherein the method comprises the following steps: photographing an electric wiring field based on APP software on the 5G mobile terminal; the distributed cloud storage server stores the electric wiring scene photos; the distributed cloud storage server carries out image preprocessing on the electric wiring field photo; identifying whether the electrical wiring in the electrical wiring digital image is correct based on the trained convolutional neural network; generating electrical inspection information based on the convolutional neural network recognition result, the electrical inspection information being used to guide electrical installers to make correct wiring. The invention realizes the quick uploading of the electric wiring photos and the quick feedback of the cloud analysis results, improves the judgment speed of the electric wiring correctness and mistakes, reduces the field wiring operation time and improves the working efficiency.
Description
Technical Field
The invention mainly relates to the technical field of computers, in particular to a method and a system for automatically checking the quality of electric manual wiring on line.
Background
The correctness of electrical wiring depends on the work responsibility, the level of business and the work proficiency of the wiring personnel. Abnormal conditions such as misconnection, disconnection and the like cause additional line loss or unstable working conditions of the electrical equipment, so that the electric quantity of the electrical equipment is lost or damaged, and the economic loss is suffered by users. Generally, there are a variety of obvious wiring characteristics that can be used for wiring misjudgment. For example, the color of the wire can be used for judgment, such as the zero line is blue, and the live line is red; the wire diameter can be used for judgment, and generally, the wire diameter of a high-current wire is thicker, and the wire diameter of a low-current wire is thinner.
The characteristics are identified by means of an artificial intelligence method, and the identification speed and accuracy are greatly improved. However, the quality inspection of the manual electric wiring based on the manual inspection method is long in time consumption and low in working efficiency; the quality inspection of manual electrical wiring is carried out based on a manual inspection mode, and the requirement on the professional level of workers is high.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides an online automatic inspection method and system for the electrical manual wiring quality, and provides the online automatic inspection method and system for the electrical manual wiring quality based on a convolutional neural network, so that the problems of long time consumption, low working efficiency, high requirement on the professional level of personnel and the like in the electrical manual wiring quality inspection are solved.
The invention provides an on-line automatic inspection method for the quality of electric manual wiring, which comprises the following steps:
photographing an electrical wiring field based on APP software on the 5G mobile terminal to obtain a picture of the electrical wiring field;
sending the electric wiring field photo to a distributed cloud storage server based on the mapping address of the APP software, wherein the distributed cloud storage server stores the electric wiring field photo;
the distributed cloud storage server carries out image preprocessing on the electric wiring field photo to generate an electric wiring digital image which accords with the prediction processing of a convolutional neural network;
identifying whether the electrical wiring in the electrical wiring digital image is correct based on the trained convolutional neural network;
generating electrical inspection information based on the convolutional neural network recognition result, the electrical inspection information being used to guide electrical installers to make correct wiring.
Multiple cloud servers in the distributed cloud storage server deploy the same network operating system and Hadoop distributed file processing system.
The distributed cloud storage server image preprocessing of the electric wiring field photo comprises the following steps:
format converting the electrical wiring site picture to generate a first pre-processed image,
performing edge clipping on the first preprocessed image to generate a second preprocessed image;
carrying out size transformation on the second preprocessed image to generate a third preprocessed image;
and performing brightness adjustment on the third preprocessed image to generate an electrical wiring digital image which is in accordance with the prediction processing of the convolutional neural network.
The training-based convolutional neural network identifying whether the electrical wiring in the electrical wiring digital image is correct comprises:
performing data slicing on the electrical wiring digital image based on a sliding window, and identifying wiring characteristics in the electrical wiring digital image;
inputting the recognized wiring characteristics into a convolutional neural network to fit the electrical wiring digital image to extract characteristic vectors;
carrying out batch normalization processing on the fitted characteristic vectors;
carrying out global average pooling on the feature vectors subjected to batch normalization processing to obtain a feature map of a specified output dimension;
processing data of the characteristic diagram through a Dropout layer of a neural network, and inhibiting overfitting;
inputting the processed image data stream into a softmax layer for operation to judge whether the electrical wiring in the electrical wiring digital image is correct or not.
The wiring features include: line color, line diameter, line number, and cross relationship.
The step of inputting the identified wiring features into a convolutional neural network to fit the electrical wiring digital image and extract feature vectors comprises the following steps:
inputting a historical electrical wiring digital image set into a first-stage convolution network of a convolution neural network, and performing first-stage fitting processing on the electrical wiring digital image based on the first-stage convolution network, wherein the first-stage convolution network comprises three layers of convolution networks, each layer of convolution network is respectively provided with 16 convolution kernels with 11 pixels multiplied by 11 pixels, 48 convolution kernels with 5 pixels multiplied by 5 pixels, 52 convolution kernels with 3 pixels multiplied by 3 pixels, and the last layer is an average pooling layer;
inputting the historical electrical wiring digital image set subjected to the first-stage fitting processing into a second-stage convolution network of a convolution neural network, and performing second-stage fitting processing on the electrical wiring digital image based on the second-stage convolution network, wherein the second-stage convolution network comprises three layers of convolution networks, each layer of convolution network is respectively provided with 64 convolution kernels of 3 pixels multiplied by 3 pixels, 64 convolution kernels of 3 pixels multiplied by 3 pixels and 64 convolution kernels of 5 pixels multiplied by 5 pixels, and the last layer is an average pooling layer;
inputting the historical electrical wiring digital image set after the second-level fitting processing into a third-level full-connection network of a convolutional neural network, and performing third-level fitting processing on the electrical wiring digital image based on the third-level full-connection network, wherein the third-level full-connection network comprises a three-layer network, 96 neurons and 10 neurons are respectively arranged on each layer of the three-layer network, and the last layer is a softmax output layer and is used for outputting prediction probability values corresponding to 10 categories, including the prediction probability values of 1 correct wiring type and 9 wrong wiring types. The method further comprises the following steps:
and sending the electrical inspection information to the 5G mobile terminal, wherein the 5G mobile terminal displays the electrical inspection information visually based on APP software.
And when the electrical inspection information is guiding electrical wiring information, presenting the electrical inspection information to a user based on a visual window, a visual message, a visual curve graph or a visual table.
Correspondingly, the invention also provides an online automatic inspection system for the quality of the electrical manual wiring, which comprises:
the 5G mobile terminal is used for providing APP software for a user to photograph an electric wiring field and obtain a picture of the electric wiring field; sending the electric wiring field photo to a distributed cloud storage server based on the mapping address of the APP software;
the distributed cloud storage server is used for storing the electric wiring field photos; carrying out image preprocessing on the electric wiring field picture to generate an electric wiring digital image which accords with the prediction processing of a convolutional neural network; identifying whether the electrical wiring in the electrical wiring digital image is correct based on the trained convolutional neural network; generating electrical inspection information based on the convolutional neural network recognition result, the electrical inspection information being used to guide electrical installers to make correct wiring.
And the 5G mobile terminal is also used for receiving electrical inspection information and visually displaying the electrical inspection information based on APP software.
The embodiment of the invention has the following beneficial effects:
(1) by means of the advantages of low time delay and high reliability of the 5G technology, the electric wiring photos are uploaded quickly, the cloud analysis results are fed back quickly, the electric wiring correct and incorrect judgment speed is increased, the field wiring operation time is shortened, and the working efficiency is improved.
(2) The electric wiring photos and related information (such as the acquired time, place, personnel, type and the like) are centrally stored, processed and displayed on the cloud platform, so that the electronization, informatization and networking of the work flow are realized, and the work traces and the work records are left for the subsequent work spot check, quality control and operation tracing.
(3) The convolution neural network deployed on the cloud platform automatically performs feature extraction and correct and wrong judgment on the electric wiring pictures uploaded by the mobile phone APP, the judgment speed and the judgment accuracy are greatly improved compared with a manual mode, equipment damage caused by electric wiring errors is effectively avoided, and the safety and the reliability of electric equipment are enhanced.
(4) The real-time passback and intelligent analysis of the electric wiring photos improve the intelligent level of field wiring operation, reduce the burden of basic teams and groups and improve the working efficiency. The historical photos of the electrical wiring recorded by the cloud platform are convenient for a company management layer to master the working time and place, the task quantity and the error rate of teams and employees, and provide quantitative data support for performance assessment and evaluation priority evaluation of the employees.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic diagram of a system for automatically inspecting the quality of electrical manual wiring on line in an embodiment of the invention;
fig. 2 is a flow chart of a method for online automatic inspection of electrical manual wiring quality in an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention relates to an on-line automatic inspection method for the quality of electric manual wiring, which comprises the following steps: photographing an electrical wiring field based on APP software on the 5G mobile terminal to obtain a picture of the electrical wiring field; sending the electric wiring field photo to a distributed cloud storage server based on the mapping address of the APP software, wherein the distributed cloud storage server stores the electric wiring field photo; the distributed cloud storage server carries out image preprocessing on the electric wiring field photo to generate an electric wiring digital image which accords with the prediction processing of a convolutional neural network; identifying whether the electrical wiring in the electrical wiring digital image is correct based on the trained convolutional neural network; generating electrical inspection information based on the convolutional neural network recognition result, the electrical inspection information being used to guide electrical installers to make correct wiring.
Specifically, fig. 1 shows a schematic structural diagram of a system for online automatic inspection of electrical manual wiring quality in an embodiment of the present invention, where the system includes:
the 5G mobile terminal is used for providing APP software for a user to photograph an electric wiring field and obtain a picture of the electric wiring field; sending the electric wiring field photo to a distributed cloud storage server based on the mapping address of the APP software;
the distributed cloud storage server is used for storing the electric wiring field photos; carrying out image preprocessing on the electric wiring field picture to generate an electric wiring digital image which accords with the prediction processing of a convolutional neural network; identifying whether the electrical wiring in the electrical wiring digital image is correct based on the trained convolutional neural network; generating electrical inspection information based on the convolutional neural network recognition result, the electrical inspection information being used to guide electrical installers to make correct wiring.
And the 5G mobile terminal is also used for receiving electrical inspection information and visually displaying the electrical inspection information based on APP software.
According to the embodiment of the invention, an electric wiring intelligent inspection system is formed by means of 5G infrastructure such as a 5G public network, a 5G smart phone, a cloud platform and a 5G switch and matched software, so that the quick and reliable transmission of electric wiring pictures and data is realized, and the efficient fusion of personnel, equipment and information is realized. The embodiment of the invention fully utilizes the advantages of new-generation information technologies such as 5G, big data, artificial intelligence and the like, takes an electric wiring picture based on the mobile phone APP which is developed autonomously, then automatically uploads the picture to the cloud platform through the 5G network, performs wiring characteristic recognition on the picture by utilizing a convolutional neural network prediction model on the cloud platform, accurately and efficiently judges whether the electric wiring is correct or not, and transmits the judgment result back to a visual interface of the mobile phone APP through the 5G network for displaying, thereby prompting and guiding field personnel to finish the wiring correctly.
It should be noted that the hardware of the system includes: support smart mobile phone, 5G network, 5G switch and the cloud ware of 5G network, the software of system includes: the system comprises client APP software installed in a smart phone, a network operating system installed in a cloud server, a distributed file processing system and electrical manual wiring quality inspection software. The 5G network is a 5G mobile phone communication network of a telecom operator, and the network can well cover an electric wiring field. The 5G switch works on a physical layer and a data link layer and is used for data exchange and forwarding between the 5G network and the plurality of cloud servers. The same network operating system and the same Hadoop distributed file processing system are deployed in the cloud servers, and electric manual wiring quality inspection software is deployed. The client APP software is used for uploading the electric wiring photos to the cloud server in real time and performing intelligent analysis, and the intelligent level of field wiring operation is improved. The electric manual wiring quality inspection software is developed based on the principle of a convolutional neural network, the convolutional neural network is automatically updated by adopting an online learning mechanism, and more wiring error types are learned by sequentially adding new pictures in operation, so that the accuracy of judgment is continuously improved.
Specifically, fig. 2 shows a flowchart of a method for automatically inspecting the quality of the electrical manual wiring on line in the embodiment of the present invention, which includes the following steps:
s201, photographing an electric wiring field based on APP software on a 5G mobile terminal to obtain a picture of the electric wiring field;
s202, sending the electric wiring field photo to a distributed cloud storage server based on a mapping address of APP software, wherein the distributed cloud storage server stores the electric wiring field photo;
it should be noted that multiple cloud servers in the distributed cloud storage server deploy the same network operating system and Hadoop distributed file processing system.
It should be noted that the distributed cloud storage server organically concentrates the electrical wiring photos of different sources, formats and sizes logically or physically, and provides comprehensive historical data storage, sharing and query services, so as to facilitate deep analysis and mining of large data at a later stage.
S203, the distributed cloud storage server performs image preprocessing on the electric wiring field picture to generate an electric wiring digital image which accords with the prediction processing of a convolutional neural network;
it should be noted that the image preprocessing of the electric wiring scene photo by the distributed cloud storage server includes: format conversion is carried out on the picture of the electric wiring site to generate a first preprocessing image, and edge cutting is carried out on the first preprocessing image to generate a second preprocessing image; carrying out size transformation on the second preprocessed image to generate a third preprocessed image; and performing brightness adjustment on the third preprocessed image to generate an electrical wiring digital image which is in accordance with the prediction processing of the convolutional neural network.
S204, identifying whether the electrical wiring in the electrical wiring digital image is correct or not based on the trained convolutional neural network;
it should be noted that the input image of the convolutional neural network is an RGB color image of 227 pixels × 227 pixels, and the output result is to output prediction probability values corresponding to 10 categories, that is, prediction probability values of 1 correct wiring type and 9 wrong wiring types. The input image of the convolution neural network is an RGB color image of 227 pixels x 227 pixels, and the activation function of the convolution neural network is ReLu.
It should be noted that, here, identifying whether the electrical wiring in the electrical wiring digital image is correct based on the trained convolutional neural network includes: performing data slicing on the electrical wiring digital image based on a sliding window, and identifying wiring characteristics in the electrical wiring digital image; inputting the recognized wiring characteristics into a convolutional neural network to fit the electrical wiring digital image to extract characteristic vectors; carrying out batch normalization processing on the fitted characteristic vectors; carrying out global average pooling on the feature vectors subjected to batch normalization processing to obtain a feature map of a specified output dimension; processing data of the characteristic diagram through a Dropout layer of a neural network, and inhibiting overfitting; inputting the processed image data stream into a softmax layer for operation to judge whether the electrical wiring in the electrical wiring digital image is correct or not.
It should be noted that the connection feature includes: line color, line diameter, line number, and cross relationship.
It should be noted that, inputting the identified wiring features into the convolutional neural network to fit the electrical wiring digital image and extract feature vectors includes:
inputting a historical electric wiring digital image set into a first-stage convolution network of a convolution neural network, and performing first-stage fitting processing on the electric wiring digital image based on the first-stage convolution network, wherein the first-stage convolution network comprises three layers of convolution networks, each layer of convolution network is respectively provided with 16 convolution kernels of 11 pixels multiplied by 11 pixels, 48 convolution kernels of 5 pixels multiplied by 5 pixels, and 52 convolution kernels of 3 pixels multiplied by 3 pixels, and the last layer is an average pooling layer;
inputting the historical electric wiring digital image set subjected to the first-stage fitting processing into a second-stage convolution network of a convolution neural network, and performing second-stage fitting processing on the electric wiring digital image based on the second-stage convolution network, wherein the second-stage convolution network comprises three layers of convolution networks, each layer of convolution network is respectively provided with 64 convolution kernels of 3 pixels multiplied by 3 pixels, 64 convolution kernels of 3 pixels multiplied by 3 pixels and 64 convolution kernels of 5 pixels multiplied by 5 pixels, and the last layer is an average pooling layer;
the third level fully-connected network of the convolutional neural network is input with the historical electric wiring digital image set after the second level fitting processing, and is right based on the third level fully-connected network the electric wiring digital image carries out the third level fitting processing, the third level fully-connected network, including three layer network, be provided with 96, 10 neurons on each layer network respectively, last one deck is softmax output layer for the prediction probability value that output 10 categories correspond, including the prediction probability value of 1 kind correct wiring type and 9 kinds of wrong wiring types. Here, the batch normalization processing on the fitted feature vectors includes: calculating the mean value of the fitted historical electrical wiring digital image set; calculating the variance of the fitted historical electrical wiring digital image set; normalizing the fitted historical electrical wiring digital image set by using the calculated mean value and variance to obtain 0-1 distribution; and carrying out scale transformation and offset processing on the fitted historical electrical wiring digital image set. The mean value and the variance of the fitted feature vectors are normalized, so that the convergence speed of the whole network model is increased, the problem of gradient dispersion in a deep network is relieved to a certain extent, and the deep network model is easier and more stable to train.
S205, generating electrical inspection information based on the recognition result of the convolutional neural network, wherein the electrical inspection information is used for guiding electrical installation personnel to carry out correct wiring;
s206, sending the electrical inspection information to the 5G mobile terminal, and carrying out visual display on the electrical inspection information by the 5G mobile terminal based on APP software.
And when the electrical inspection information is guiding electrical wiring information, presenting the electrical inspection information to a user based on a visual window, a visual message, a visual curve graph or a visual table. By means of the advantages of low time delay and high reliability of the 5G technology, the embodiment of the invention realizes quick uploading of the electric wiring pictures and quick feedback of the cloud analysis result, improves the speed of judging whether the electric wiring is correct or incorrect, reduces the field wiring operation time and improves the working efficiency.
The electric wiring photos and the related information (such as the acquired time, the acquired place, the acquired personnel, the acquired type and the like) are centrally stored, processed and displayed on the cloud platform, so that the electronization, informatization and networking of the work flow are realized, and the work traces and the work records are left for the subsequent work spot check, the quality control and the operation tracing.
The convolutional neural network deployed on the cloud platform in the embodiment of the invention automatically performs feature extraction and correct and wrong judgment on the electric wiring picture uploaded by the mobile phone APP, so that the judgment speed and accuracy are greatly improved compared with those of a manual mode, the equipment damage caused by electric wiring errors is effectively avoided, and the safety and reliability of electric equipment are enhanced.
The real-time return and intelligent analysis of the electric wiring photos in the embodiment of the invention improve the intelligent level of field wiring operation, reduce the burden of basic teams and groups and improve the working efficiency. The historical photos of the electrical wiring recorded by the cloud platform are convenient for a company management layer to master the working time and place, the task quantity and the error rate of teams and employees, and provide quantitative data support for performance assessment and evaluation priority evaluation of the employees.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable storage medium, and the storage medium may include: a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic or optical disk, or the like.
In addition, the above embodiments of the present invention are described in detail, and the principle and the implementation manner of the present invention should be described herein by using specific examples, and the above description of the embodiments is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. An on-line automatic inspection method for the quality of electric manual wiring is characterized by comprising the following steps:
photographing an electrical wiring field based on APP software on the 5G mobile terminal to obtain a picture of the electrical wiring field;
sending the electric wiring field photo to a distributed cloud storage server based on the mapping address of the APP software, wherein the distributed cloud storage server stores the electric wiring field photo;
the distributed cloud storage server carries out image preprocessing on the electric wiring field photo to generate an electric wiring digital image which accords with the prediction processing of a convolutional neural network;
identifying whether the electrical wiring in the electrical wiring digital image is correct based on the trained convolutional neural network;
generating electrical inspection information based on the convolutional neural network recognition result, the electrical inspection information being used to guide electrical installers to make correct wiring.
2. The method for the online automatic inspection of the quality of the electrical manual wiring according to claim 1, wherein a plurality of cloud servers in the distributed cloud storage server deploy the same network operating system and Hadoop distributed file processing system.
3. The method for the online automatic inspection of the quality of the electrical manual wiring according to claim 1, wherein the image preprocessing of the electrical wiring scene photo by the distributed cloud storage server comprises:
format converting the electrical wiring site picture to generate a first pre-processed image,
performing edge clipping on the first preprocessed image to generate a second preprocessed image;
carrying out size transformation on the second preprocessed image to generate a third preprocessed image;
and performing brightness adjustment on the third preprocessed image to generate an electrical wiring digital image which is in accordance with the prediction processing of the convolutional neural network.
4. The method of online automated inspection of electrical manual wire quality of claim 1, wherein the training-based convolutional neural network identifying whether electrical wires in the electrical wire digital image are correct comprises:
performing data slicing on the electrical wiring digital image based on a sliding window, and identifying wiring characteristics in the electrical wiring digital image;
inputting the recognized wiring characteristics into a convolutional neural network to fit the electrical wiring digital image to extract characteristic vectors;
carrying out batch normalization processing on the fitted characteristic vectors;
carrying out global average pooling on the feature vectors subjected to batch normalization processing to obtain a feature map of a specified output dimension;
processing data of the characteristic diagram through a Dropout layer of a neural network, and inhibiting overfitting;
inputting the processed image data stream into a softmax layer for operation to judge whether the electrical wiring in the electrical wiring digital image is correct or not.
5. The method for the on-line automatic inspection of the quality of electrical manual wiring according to claim 4, wherein the wiring characteristics include: line color, line diameter, line number, and cross relationship.
6. The method of online automated inspection of electrical manual wire quality of claim 4, wherein said inputting the identified wire characteristics into a convolutional neural network to fit the electrical wire digital image to extract a feature vector comprises:
inputting a historical electrical wiring digital image set into a first-stage convolution network of a convolution neural network, and performing first-stage fitting processing on the electrical wiring digital image based on the first-stage convolution network, wherein the first-stage convolution network comprises three layers of convolution networks, each layer of convolution network is respectively provided with 16 convolution kernels with 11 pixels multiplied by 11 pixels, 48 convolution kernels with 5 pixels multiplied by 5 pixels, 52 convolution kernels with 3 pixels multiplied by 3 pixels, and the last layer is an average pooling layer;
inputting the historical electrical wiring digital image set subjected to the first-stage fitting processing into a second-stage convolution network of a convolution neural network, and performing second-stage fitting processing on the electrical wiring digital image based on the second-stage convolution network, wherein the second-stage convolution network comprises three layers of convolution networks, each layer of convolution network is respectively provided with 64 convolution kernels of 3 pixels multiplied by 3 pixels, 64 convolution kernels of 3 pixels multiplied by 3 pixels and 64 convolution kernels of 5 pixels multiplied by 5 pixels, and the last layer is an average pooling layer;
the third level fully-connected network of the convolutional neural network is input with the historical electric wiring digital image set after the second level fitting processing, and is right based on the third level fully-connected network the electric wiring digital image carries out the third level fitting processing, the third level fully-connected network, including three layer network, be provided with 96, 10 neurons on each layer network respectively, last one deck is softmax output layer for the prediction probability value that output 10 categories correspond, including the prediction probability value of 1 kind correct wiring type and 9 kinds of wrong wiring types.
7. The method for the on-line automatic inspection of the quality of the electrical manual wiring according to any one of claims 1 to 6, characterized in that it further comprises, after said method:
and sending the electrical inspection information to the 5G mobile terminal, wherein the 5G mobile terminal displays the electrical inspection information visually based on APP software.
8. The method for the online automatic inspection of the quality of the electrical manual wiring according to claim 7, wherein when the electrical inspection information is guiding electrical wiring information, the electrical inspection information is presented to a user based on a visual window, a visual message, a visual graph or a visual table.
9. An on-line automatic inspection system for electrical manual wiring quality, the system comprising:
the 5G mobile terminal is used for providing APP software for a user to photograph an electric wiring field and obtain a picture of the electric wiring field; sending the electric wiring field photo to a distributed cloud storage server based on the mapping address of the APP software;
the distributed cloud storage server is used for storing the electric wiring field photos; carrying out image preprocessing on the electric wiring field picture to generate an electric wiring digital image which accords with the prediction processing of a convolutional neural network; identifying whether the electrical wiring in the electrical wiring digital image is correct based on the trained convolutional neural network; generating electrical inspection information based on the convolutional neural network recognition result, the electrical inspection information being used to guide electrical installers to make correct wiring.
10. The system for the online automatic inspection of the quality of the electrical manual wiring according to claim 9, wherein the 5G mobile terminal is further configured to receive electrical inspection information and visually display the electrical inspection information based on APP software.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210442216.8A CN114979184A (en) | 2022-04-25 | 2022-04-25 | Method and system for online automatic inspection of electrical manual wiring quality |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210442216.8A CN114979184A (en) | 2022-04-25 | 2022-04-25 | Method and system for online automatic inspection of electrical manual wiring quality |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114979184A true CN114979184A (en) | 2022-08-30 |
Family
ID=82980407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210442216.8A Pending CN114979184A (en) | 2022-04-25 | 2022-04-25 | Method and system for online automatic inspection of electrical manual wiring quality |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114979184A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112287773A (en) * | 2020-10-10 | 2021-01-29 | 国家电网有限公司 | Primary wiring diagram primitive identification method based on convolutional neural network |
| CN112633343A (en) * | 2020-12-16 | 2021-04-09 | 国网江苏省电力有限公司检修分公司 | Power equipment terminal strip wiring checking method and device |
| CN112784853A (en) * | 2020-12-24 | 2021-05-11 | 深兰人工智能芯片研究院（江苏）有限公司 | Terminal connection state detection method and device |
-
2022
- 2022-04-25 CN CN202210442216.8A patent/CN114979184A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112287773A (en) * | 2020-10-10 | 2021-01-29 | 国家电网有限公司 | Primary wiring diagram primitive identification method based on convolutional neural network |
| CN112633343A (en) * | 2020-12-16 | 2021-04-09 | 国网江苏省电力有限公司检修分公司 | Power equipment terminal strip wiring checking method and device |
| CN112784853A (en) * | 2020-12-24 | 2021-05-11 | 深兰人工智能芯片研究院（江苏）有限公司 | Terminal connection state detection method and device |
Non-Patent Citations (1)
| Title |
|---|
| 徐振磊等: "基于图像识别技术的输电线路智能监控系统应用", 计算机系统应用, no. 01, pages 2 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114362367B (en) | Cloud-edge-cooperation-oriented power transmission line monitoring system and method, and cloud-edge-cooperation-oriented power transmission line identification system and method | |
| CN107330731B (en) | Method and device for identifying click abnormity of advertisement space | |
| US20220222796A1 (en) | Image processing method and apparatus, server, and storage medium | |
| CN109308490A (en) | Method and apparatus for generating information | |
| CN111414842B (en) | Video comparison method and device, computer equipment and storage medium | |
| CN112580657B (en) | Self-learning character recognition method | |
| CN111611351A (en) | Control method and device for online customer service session and electronic equipment | |
| CN114091472B (en) | Training method of multi-label classification model | |
| CN112800090A (en) | Data processing method combining edge computing and path analysis and big data cloud platform | |
| CN113657207B (en) | Cloud-side cooperative power distribution station fire light intelligent monitoring method and system | |
| CN114913376A (en) | Image-based defect automatic identification method, device and system and storage medium | |
| CN112200465B (en) | Electric power AI method and system based on multimedia information intelligent analysis | |
| CN111898544B (en) | Text image matching method, device and equipment and computer storage medium | |
| CN116580362B (en) | Transmission operation cross-system fusion data acquisition method and digital asset processing system | |
| CN114979184A (en) | Method and system for online automatic inspection of electrical manual wiring quality | |
| CN115879880A (en) | RPA (resilient packet Access) acquisition method and system for power industry expansion data | |
| CN113743129B (en) | Information pushing method, system, equipment and medium based on neural network | |
| CN113362083B (en) | Repair reporting method and device, electronic equipment and storage medium | |
| CN115756256A (en) | Information labeling method, system, electronic equipment and storage medium | |
| CN113298112B (en) | Integrated data intelligent labeling method and system | |
| CN115204841A (en) | Workflow generation method and device, electronic equipment and storage medium | |
| CN108957221A (en) | A kind of transmission line of electricity defect early warning system based on web | |
| CN109740858A (en) | Automation aid decision-making system and method based on deep learning | |
| CN115102852B (en) | Internet of things service opening method and device, electronic equipment and computer medium | |
| CN118279664A (en) | Identification method and system for mountain fire of corridor 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 |