CN114913172B - Method, system, equipment and medium for identifying manufacturing risk of cable middle head - Google Patents
Method, system, equipment and medium for identifying manufacturing risk of cable middle head Download PDFInfo
- Publication number
- CN114913172B CN114913172B CN202210818638.0A CN202210818638A CN114913172B CN 114913172 B CN114913172 B CN 114913172B CN 202210818638 A CN202210818638 A CN 202210818638A CN 114913172 B CN114913172 B CN 114913172B
- Authority
- CN
- China
- Prior art keywords
- risk
- video stream
- preset
- identification model
- information
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a method, a system, equipment and a medium for identifying the manufacturing risk of a cable middle head. When the video stream characteristics meet preset production starting conditions, identifying the risk types to which the video stream characteristics belong through a risk identification model, constructing risk information by adopting the risk types, counting the constructed quantity of the risk information in real time, and when the constructed quantity is larger than or equal to a preset risk threshold value, sending a risk warning and resetting the constructed quantity. When the video stream characteristics meet the preset manufacturing completion conditions, the corresponding working risk report is generated by adopting the existing all risk information, the video stream characteristics of the real-time video stream are extracted by adopting the trained risk identification model, and the characteristics are subjected to risk identification, so that the real-time risk monitoring is realized, the identification speed is increased, and the accuracy of the identification result is high.
Description
Technical Field
The invention relates to the technical field of cable middle head manufacturing risk identification, in particular to a method, a system, equipment and a medium for identifying cable middle head manufacturing risks.
Background
With the continuous development of power technology, many safety risks still exist when power operators operate equipment and work on the spot. In order to further guarantee the safe production of electric power, besides basic monitoring measures are arranged on site, a remote video monitoring mode is added.
At present, in the process of manufacturing the thermal shrinkage cable intermediate head, a common remote video monitoring mode is manual video monitoring, video information of the process of manufacturing the thermal shrinkage cable intermediate head is obtained in real time through monitoring equipment such as a camera and the like, and is transmitted to a monitoring terminal to be remotely monitored in real time by a supervisor.
Although the electric power safety production capacity is improved by the monitoring mode, more human resources are needed for supervision, and the manual supervision is easy to cause monitoring negligence, so that the monitoring accuracy is low.
Disclosure of Invention
The invention provides a method, a system, equipment and a medium for identifying manufacturing risks of a cable middle head, and solves the technical problems that the conventional cable middle head manufacturing monitoring mode improves the power safety production capacity, but needs to spend more human resources for monitoring, and is easy to cause monitoring negligence due to manual monitoring, so that the monitoring accuracy is low.
The invention provides a method for identifying the manufacturing risk of a cable middle head, which comprises the following steps:
acquiring a real-time video stream between cable middle head production;
extracting video stream characteristics of the real-time video stream through a preset risk identification model;
when the video stream characteristics meet preset production starting conditions, identifying the risk types to which the video stream characteristics belong through the risk identification model;
adopting the risk types to construct risk information, and counting the constructed quantity of the risk information in real time;
when the construction number is greater than or equal to a preset risk threshold value, sending a risk warning and resetting the construction number;
and when the video stream characteristics meet preset manufacturing completion conditions, generating a corresponding working risk report by adopting all the existing risk information.
Optionally, the method further comprises:
acquiring monitoring videos made by a plurality of cable intermediate heads;
dividing each monitoring video into a plurality of static pictures according to a preset division interval to obtain a corresponding static picture set;
extracting risk pictures which accord with a preset initial risk type in the static picture set, and constructing a risk training set and a risk verification set;
and training an initial risk identification model by adopting the risk training set, and constructing the risk identification model by combining the risk verification set and a preset training standard.
Optionally, the step of training an initial risk identification model by using the risk training set, and building the risk identification model by combining the risk verification set and a preset training standard includes:
amplifying the number of the risk pictures in the risk training set according to a preset picture amplification mode to obtain a target risk training set;
training an initial risk identification model by adopting the target risk training set to obtain an intermediate risk identification model;
verifying the intermediate risk identification model by adopting the risk verification set to obtain a verification result;
if the verification result meets a preset training standard, determining the intermediate risk identification model as the risk identification model;
and if the verification result does not meet the preset training standard, skipping to execute the step of training the initial risk identification model by adopting the target risk training set to obtain an intermediate risk identification model.
Optionally, the risk identification model includes a convolutional layer; the step of extracting the video stream features of the real-time video stream through a preset risk identification model comprises the following steps:
extracting video frame pictures of the real-time video stream according to a preset video extraction interval;
and extracting the video stream characteristics of each video frame picture through the convolution layer.
Optionally, the risk identification model comprises a risk comparison module; when the video stream features meet preset production starting conditions, the step of identifying the risk types to which the video stream features belong through the risk identification model comprises the following steps:
when the video stream characteristics meet preset production starting conditions, risk similarity between the video stream characteristics and a plurality of preset risk characteristics is calculated through the risk comparison module;
and selecting the risk type to which the risk feature corresponding to the maximum risk similarity belongs as the risk type to which the video stream feature belongs.
Optionally, the step of constructing risk information by using the risk types and counting the constructed quantity of the risk information in real time includes:
acquiring a risk keyword carried by the risk type;
extracting a safety risk point, risk occurrence time and a risk picture corresponding to the risk keyword from a preset video stream characteristic information base;
and constructing risk information by adopting the safety risk points, the risk occurrence time and the risk pictures, and counting the constructed quantity of the risk information in real time.
Optionally, the risk identification model comprises an end feature comparison module; when the video stream characteristics meet preset manufacturing completion conditions, the step of generating the corresponding working risk report by adopting all the existing risk information comprises the following steps:
calculating the feature similarity between the video stream features and preset production ending features through the ending feature comparison module;
when the feature similarity meets a preset threshold, respectively loading corresponding page components by adopting all the existing risk information;
rendering all the page components to generate a work risk report.
The invention also provides a system for identifying the manufacturing risk of the cable middle head, which comprises the following components:
the real-time video stream acquisition module is used for acquiring a real-time video stream between cable middle head production;
the video stream feature extraction module is used for extracting video stream features of the real-time video stream through a preset risk identification model;
the risk type identification module is used for identifying the risk type to which the video stream characteristics belong through the risk identification model when the video stream characteristics meet preset production starting conditions;
the risk information construction and construction quantity counting module is used for constructing risk information by adopting the risk types and counting the construction quantity of the risk information in real time;
the risk warning sending and construction number resetting module is used for sending a risk warning and resetting the construction number when the construction number is larger than or equal to a preset risk threshold;
and the working risk report generating module is used for generating a corresponding working risk report by adopting the existing risk information when the video stream characteristics meet the preset manufacturing completion conditions.
The invention also provides an electronic device, which comprises a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of implementing the method for identifying the manufacturing risk of the cable middle head.
The invention also provides a computer-readable storage medium, on which a computer program is stored, which, when executed, implements a method for identifying a risk of making a cable intermediate head as in any one of the above.
According to the technical scheme, the invention has the following advantages:
the method comprises the steps of obtaining a real-time video stream between cable middle head production and extracting video stream characteristics of the real-time video stream through a preset risk identification model. When the video stream characteristics meet the preset making starting conditions, the risk types to which the video stream characteristics belong are identified through the risk identification model, risk information is constructed by adopting the risk types, the constructed quantity of the risk information is counted in real time, and when the constructed quantity is larger than or equal to the preset risk threshold value, a risk warning is sent and the constructed quantity is reset. When the video stream characteristics meet the preset manufacturing completion conditions, all existing risk information is adopted to generate corresponding working risk reports, and the technical problems that although the electric power safety production capacity is improved, more human resources are needed to be spent for supervision, and manual supervision is easy to cause monitoring negligence, so that the monitoring accuracy is low are solved. The trained risk recognition model is adopted to extract the video stream characteristics of the real-time video stream, and risk recognition is carried out on the characteristics, so that real-time risk monitoring is realized, the recognition speed is increased, and the recognition result accuracy is high.
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, and 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 these drawings without inventive exercise.
Fig. 1 is a flowchart illustrating steps of a method for identifying a manufacturing risk of a cable intermediate head according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating steps of a method for identifying a manufacturing risk of a cable middle head according to a second embodiment of the present invention;
FIG. 3 is a block diagram of a process for training a risk recognition model according to a second embodiment of the present invention;
fig. 4 is a flowchart of a method for identifying a manufacturing risk of a cable middle head according to a second embodiment of the present invention;
fig. 5 is a block diagram of a system for identifying a risk of manufacturing a cable intermediate head according to a third embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a method, a system, equipment and a medium for identifying manufacturing risks of a cable middle head, which are used for solving the technical problems that although the electric power safety production capacity is improved, the conventional cable middle head manufacturing monitoring mode needs to spend more human resources for monitoring, and the monitoring is easy to neglect due to manual monitoring, so that the monitoring accuracy is low.
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. 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.
Referring to fig. 1, fig. 1 is a flowchart illustrating steps of a method for identifying a manufacturing risk of a cable middle head according to an embodiment of the present invention.
The invention provides a method for identifying the manufacturing risk of a cable middle head, which comprises the following steps:
The cable middle head manufacturing room refers to a working environment for manufacturing the cable middle head, and can be an outdoor environment, an indoor environment or a cable tunnel and the like. The cable middle head is usually made by connecting two cables by using a heat shrinkage method. The real-time video streaming refers to the real-time transmission and acquisition of video data in the cable middle head manufacturing process by video monitoring equipment arranged in a cable middle head manufacturing room.
In the embodiment of the invention, the video monitoring equipment arranged between the cable middle head manufacturing rooms is obtained to transmit the collected video data of the cable middle head manufacturing process in real time.
And 102, extracting video stream characteristics of the real-time video stream through a preset risk identification model.
The risk identification model is a neural network model which can extract video stream features of the video stream and identify functions such as risk types to which the video stream features belong. The video stream features refer to picture features including color features, texture features, shape features and spatial relationship features of video frame pictures.
In the embodiment of the invention, video frame pictures of the real-time video stream are extracted according to the preset video extraction interval, and the video stream characteristics of each video frame picture are extracted through the convolution layer of the risk identification model.
And 103, when the video stream characteristics meet preset making starting conditions, identifying the risk types to which the video stream characteristics belong through a risk identification model.
The preset manufacturing starting condition refers to the picture characteristics that the cable protective layer begins to be stripped and the cable is exposed out of the internal structure (the metal shielding layer, the semiconductor layer or the insulating layer).
The risk types comprise a rainy day operation risk type, a non-safety-surrounding risk type, a non-wearing safety helmet risk type, a construction lighting deficiency risk type in the cable tunnel and a spray gun illegal use risk type, and the risk characteristics corresponding to the risk types comprise a rainy day operation risk characteristic, a non-safety-surrounding risk characteristic, a non-wearing safety helmet risk characteristic, a construction lighting deficiency risk characteristic in the cable tunnel and a spray gun illegal use risk characteristic.
In the embodiment of the invention, when the video stream characteristic is that the cable sheath starts to be stripped and the cable is exposed out of the picture characteristic of the internal structure (the metal shielding layer, the semiconductor layer or the insulating layer), the risk similarity between the video stream characteristic and a plurality of preset risk characteristics is calculated through the risk comparison module of the risk identification model. And taking the risk type to which the risk feature corresponding to the maximum similarity belongs as the risk type to which the video stream feature belongs. And when the video stream characteristics do not meet the preset production starting conditions, continuously identifying whether the subsequently extracted video stream characteristics meet the preset production starting conditions.
And 104, constructing risk information by adopting the risk types, and counting the constructed quantity of the risk information in real time.
The risk information includes a security risk point, a risk occurrence time, and a risk picture. The construction quantity refers to the number of times each risk type is identified in the risk identification process, and the construction quantity can be defined as the number of times each risk type is identified in the risk identification process
、
、…、
And the value of n is the number of times of identification of each risk type.
In the embodiment of the invention, according to the risk type to which the video stream feature belongs, the safety risk point, the risk occurrence time and the risk picture corresponding to the video stream feature are determined, the safety risk point, the risk occurrence time and the risk picture are adopted to construct the risk information corresponding to the video stream feature, and the constructed quantity of the risk information is counted in real time according to the risk type.
And 105, when the construction quantity is larger than or equal to the preset risk threshold value, sending a risk warning and resetting the construction quantity.
The risk threshold is defined according to the risk type as
、
、…、
(ii) a Wherein the initial value of the construction quantity corresponding to the risk type is
The risk threshold is adjusted and set according to the risk type and the preset video extraction interval, the value of the risk threshold corresponding to each risk type may be the same or different, and the embodiment of the present invention does not limit this.
In the embodiment of the invention, when the construction quantity corresponding to any risk type is greater than or equal to the corresponding preset risk threshold, a risk warning is sent to remind a worker between cable middle head manufacturing to standardize manufacturing behaviors, the corresponding construction quantity is reset, and then whether the video stream characteristics meet the preset manufacturing completion conditions is judged. And when the construction quantity corresponding to all the risk types is smaller than the corresponding preset risk threshold value, judging whether the video stream characteristics meet the preset manufacturing completion condition or not.
And 106, when the video stream characteristics meet the preset manufacturing completion conditions, generating a corresponding work risk report by adopting all existing risk information.
The manufacturing finishing condition is the picture characteristic that the outer protective sleeve is arranged on the interfaces of the two cables and the outer protective sleeve and the cables are fully shrunk and tightened. The work risk report refers to a risk report page constructed by risk information corresponding to the risk type to which each video stream feature belongs.
In the embodiment of the invention, the feature similarity between the video stream feature and the preset manufacturing ending feature is calculated through an ending feature comparison module of the risk identification model, when the feature similarity meets a preset threshold value, corresponding page components are respectively loaded by adopting all existing risk information, all the page components are rendered, and a working risk report is generated. And when the feature similarity does not meet a preset threshold, skipping and executing the step of identifying the risk type to which the video stream feature belongs through the risk identification model.
In the embodiment of the invention, the real-time video stream between cable middle head production is obtained, and the video stream characteristics of the real-time video stream are extracted through a preset risk identification model. When the video stream characteristics meet the preset making starting conditions, the risk types to which the video stream characteristics belong are identified through the risk identification model, risk information is constructed by adopting the risk types, the constructed quantity of the risk information is counted in real time, and when the constructed quantity is larger than or equal to the preset risk threshold value, a risk warning is sent and the constructed quantity is reset. When the video stream characteristics meet the preset manufacturing completion conditions, all existing risk information is adopted to generate corresponding working risk reports, and the technical problems that although the electric power safety production capacity is improved, more human resources are needed for supervision, and monitoring negligence easily occurs when manual supervision is carried out in the existing cable middle head manufacturing monitoring mode, so that the monitoring accuracy is low are solved. The video stream characteristics of the real-time video stream are extracted through the preset risk identification model to carry out real-time risk monitoring, so that the identification speed can be increased, and the identification accuracy is high.
Referring to fig. 2, fig. 2 is a flowchart illustrating steps of a method for identifying a risk of manufacturing a cable intermediate head according to a second embodiment of the present invention.
The monitoring video refers to video data which is finished in the process of manufacturing the middle head of the cable and corresponds to the manufacturing process.
In the embodiment of the invention, a large number of monitoring videos of the manufacturing process of the thermal shrinkage cable middle head are collected.
The division interval refers to the time interval timing pair corresponding to the setting of the identification requirement of the user. The still picture is an image picture obtained by dividing the monitoring video according to a preset division interval. A still picture set refers to a picture set composed of a plurality of still pictures.
In the embodiment of the invention, each monitoring video is respectively divided into a plurality of static pictures at regular time according to the preset division interval, and all the static pictures are constructed into the static picture set.
And 203, extracting risk pictures which accord with the preset initial risk types in the static picture set, and constructing a risk training set and a risk verification set.
The initial risk types are the same as the risk types, and comprise a rainy day operation risk type, a non-safety-surrounding risk type, a non-wearing safety helmet risk type, a construction lighting deficiency risk type in the cable tunnel and a spray gun illegal use risk type, wherein the risk characteristics correspondingly arranged by the risk types are a rainy day operation risk characteristic, a non-safety-surrounding risk characteristic, a non-wearing safety helmet risk characteristic, a construction lighting deficiency risk characteristic in the cable tunnel and a spray gun illegal use risk characteristic.
The risk training set comprises each risk picture which is in the static picture set and accords with the initial risk type, and is used for training the risk picture set of the initial risk identification model. The risk verification set comprises each risk picture which accords with the initial risk type in the static picture set and is used for the risk picture set of the intermediate risk identification model.
In the embodiment of the invention, the static pictures containing the risk characteristics corresponding to the initial risk types in the static picture set are extracted, the risk marking is carried out on the static pictures according to the corresponding risk types, and the risk training set and the risk verification are constructed by adopting all the extracted static pictures.
And 204, training an initial risk identification model by adopting a risk training set, and constructing a risk identification model by combining a risk verification set and a preset training standard.
Further, step 204 may comprise the following sub-steps S11-S15:
s11, amplifying the number of the risk pictures in the risk training set according to a preset picture amplification mode to obtain a target risk training set.
The image amplification mode refers to modes of adjusting image brightness and contrast, blurring, filtering or sharpening the image and the like. The target risk training set is a set of all risk pictures obtained by a picture amplification mode.
In the embodiment of the invention, the number of the risk pictures in the risk training set is increased by adjusting the picture brightness and the contrast of each risk picture in the risk training set in a mode of blurring, filtering or sharpening the risk pictures, and a target risk training set is constructed by all the increased risk pictures and the original risk pictures.
And S12, training the initial risk identification model by adopting a target risk training set to obtain an intermediate risk identification model.
The initial risk identification model refers to an identification model which is not yet subjected to risk training by adopting a training set. The medium risk identification model is an identification model which is obtained by training a target risk training set on the basis of an initial risk identification model and can be used for carrying out risk identification on pictures.
In the embodiment of the invention, a target risk training set constructed in a picture amplification mode is input into an initial risk identification model for risk training, so that an intermediate risk identification model is obtained.
And S13, verifying the intermediate risk identification model by adopting a risk verification set to obtain a verification result.
In the embodiment of the invention, each risk picture in the risk verification set is input into an intermediate risk identification model for risk identification, a convolution layer of the intermediate risk model performs convolution operation on each risk picture in the risk verification set by using a convolution filter, and the picture characteristics are extracted so as to output corresponding risk characteristics respectively, a risk comparison module of the intermediate risk model calculates the similarity between the risk characteristics output by the convolution layer and the risk characteristics corresponding to each initial risk type by adopting a Hash algorithm, and the identification calculation result is taken as the verification result.
And S14, if the verification result meets the preset training standard, determining the intermediate risk identification model as a risk identification model.
The preset training standard refers to the accuracy requirement of the risk recognition model recognition result set according to the training requirement. The risk identification model refers to an intermediate risk identification model meeting a preset training standard.
In the embodiment of the invention, when the verification result obtained by inputting the intermediate risk identification model into the risk verification set meets the preset training standard, the intermediate risk model corresponding to the verification result is taken as the risk identification model.
And S15, if the verification result does not meet the preset training standard, skipping to execute the step of training the initial risk identification model by adopting a target risk training set to obtain an intermediate risk identification model.
In the embodiment of the invention, when the verification result obtained by inputting the intermediate risk identification model into the risk verification set does not meet the preset training standard, the training set is adopted again to train the initial risk identification model, a new intermediate risk identification model is constructed until the constructed intermediate risk identification model is verified by using the verification set, and the verification result meets the preset training standard.
As shown in fig. 3, the monitoring video produced by the cable middle head is collected, and the monitoring video is decomposed at regular time according to preset division intervals and converted into a still picture, so as to obtain a corresponding still picture set. And extracting and labeling the static pictures containing the risk characteristics corresponding to the initial risk types in the static picture set, and dividing all the extracted static pictures into a risk training set and a risk verification set. And performing data augmentation on the risk training set according to a preset image augmentation mode, and then performing risk training on the initial risk recognition model by adopting the risk training set to obtain an intermediate risk recognition model. Verifying whether the intermediate risk identification model meets a preset training standard or not by adopting a risk verification set, and if so, obtaining a risk identification model; and if the intermediate risk identification model does not meet the preset training standard, the risk training set is adopted again to train the initial risk identification model until the intermediate risk identification model is constructed and verified by using the verification set, and the verification result meets the preset training standard.
And step 205, acquiring a real-time video stream between cable middle header production.
In the embodiment of the invention, the real-time video data of the cable middle head manufacturing room, which is transmitted and collected by the video monitoring equipment arranged in the cable middle manufacturing room, is obtained.
And step 206, extracting the video stream characteristics of the real-time video stream through a preset risk identification model.
Optionally, the risk identification model includes convolutional layers, and step 206 may include the following sub-steps S21-S22:
and S21, extracting the video frame pictures of the real-time video stream according to a preset video extraction interval.
The time extraction interval refers to a time value for extracting the video frame picture at a set timing. The video frame picture refers to an image picture corresponding to a video frame included in the real-time video stream.
In the embodiment of the invention, the video frame pictures corresponding to the real-time video stream are extracted at regular time according to the preset video extraction interval.
And S22, extracting the video stream characteristics of each video frame picture through the convolution layer.
The convolutional layer is a working layer arranged on the risk identification model and used for extracting picture features.
In the embodiment of the invention, each video frame picture is input into a risk identification model, a convolution layer of the risk identification model performs convolution operation on each video frame picture by using a convolution filter, and picture characteristics are extracted, so that the corresponding video stream characteristics of each video frame picture are output.
And step 207, when the video stream characteristics meet preset making starting conditions, identifying the risk types to which the video stream characteristics belong through a risk identification model.
Optionally, the risk identification model includes a risk comparison module, and step 207 may include the following substeps S31-S32:
and S31, when the video stream characteristics meet the preset making starting conditions, respectively calculating the risk similarity between the video stream characteristics and a plurality of preset risk characteristics through a risk comparison module.
The risk characteristics comprise a rainy day operation risk characteristic, a safety enclosure non-performing risk characteristic, a safety helmet non-wearing risk characteristic, a cable tunnel internal construction lighting shortage risk characteristic and a spray gun illegal use risk characteristic, wherein the rainy day operation risk characteristic represents that raindrops appear on a picture; the risk of not carrying out safety enclosure is characterized by the absence of an enclosure device on site; the risk characteristic of the safety helmet not worn is that the head of the worker does not wear the safety helmet in the picture, and the characteristic does not include the condition that the upper part of the head is blocked by an object or the head of the worker exceeds the top of the video picture; the risk of insufficient illumination during construction in the cable tunnel is characterized in that the site is a combined condition of insufficient cable tunnel and insufficient picture light; the gun violation risk feature is characterized by a combination of gun use and presence of personnel in front of the gun.
In the embodiment of the invention, when the video stream characteristic is that the cable protective layer starts to be stripped and the cable is exposed out of the picture characteristic of the internal structure (the metal shielding layer, the semiconductor layer or the insulating layer), the risk comparison module of the risk identification model adopts a Hash algorithm to respectively calculate the similarity between the video stream characteristic and the rainy day operation risk characteristic, the non-safety-enclosure risk characteristic, the non-wearing safety helmet risk characteristic, the construction insufficient illumination risk characteristic in the cable tunnel and the spray gun illegal use risk characteristic, so as to obtain the corresponding risk similarity.
And S32, selecting the risk type to which the risk feature corresponding to the maximum risk similarity belongs as the risk type to which the video stream feature belongs.
In the embodiment of the invention, the maximum risk similarity is selected from the risk similarities between the video stream characteristics and the risk characteristics, and the risk type to which the risk characteristic corresponding to the maximum risk similarity belongs is taken as the risk type to which the video stream characteristics belong.
And step 208, adopting the risk types to construct risk information, and counting the constructed quantity of the risk information in real time.
Optionally, step 208 may include the following sub-steps S41-S43:
and S41, acquiring the risk keywords carried by the risk types.
The risk keywords refer to words bound with security risk points, risk occurrence time and risk pictures corresponding to the identified risk types in the video stream feature information base.
In the embodiment of the invention, the risk type to which the video stream feature identified by the risk identification model belongs is obtained, and the keyword carried by the risk type and corresponding to the video stream feature is obtained.
And S42, extracting a safety risk point, risk occurrence time and a risk picture corresponding to the risk keyword from a preset video stream characteristic information base.
The video stream feature information base refers to security risk points, risk occurrence time and risk pictures corresponding to each video stream feature containing the risk features.
In the embodiment of the invention, the risk keywords are input into the preset video stream characteristic information base for searching, and the safety risk points, the risk occurrence time and the risk pictures corresponding to the risk keywords are output.
And S43, adopting the safety risk points, the risk occurrence time and the risk pictures to construct risk information, and counting the constructed quantity of the risk information in real time.
The build number refers to the number of individual risk types identified during the process from the beginning of production to the completion of production.
In the embodiment of the invention, the risk information is respectively constructed by adopting the safety risk point, the risk occurrence time and the risk picture corresponding to the video stream characteristic meeting the preset risk characteristic, and the constructed quantity of the risk information is counted in real time according to the corresponding risk type when the preset risk characteristic is met each time.
And step 209, when the construction number is larger than or equal to the preset risk threshold value, sending a risk warning and resetting the construction number.
In the embodiment of the invention, if the construction quantity corresponding to any risk type is greater than or equal to the corresponding preset risk threshold, a corresponding risk warning is sent to remind workers among cable middle head manufacturing to correct manufacturing behaviors, the corresponding construction quantity is reset, and then whether the video stream characteristics meet the preset manufacturing completion conditions or not is judged. And if the construction quantity corresponding to all the risk types is smaller than the corresponding preset risk threshold value, judging whether the video stream characteristics meet preset manufacturing completion conditions or not.
And step 210, when the video stream characteristics meet the preset making completion conditions, generating a corresponding work risk report by adopting all existing risk information.
Optionally, the risk identification model includes a risk comparison module, and step 210 may include the following substeps S51-S53:
and S51, calculating the feature similarity between the video stream features and the preset production ending features through an ending feature comparison module.
In the embodiment of the invention, the video stream characteristics and the outer sheath pipe are installed on the interfaces of the two cables through the ending characteristic comparison module of the risk identification model, and the picture characteristics of the outer sheath pipe and the cables under the condition of full thermal shrinkage and tightening are compared in similarity to obtain the corresponding characteristic similarity.
And S52, when the feature similarity meets a preset threshold, respectively loading the corresponding page components by adopting all the existing risk information.
The page component refers to an interface component containing a security risk point, a risk occurrence time and a risk picture.
In the embodiment of the invention, if the feature similarity between the video stream feature and the preset production end feature meets the preset threshold, the corresponding page components are respectively loaded by adopting all the existing risk information. And if the feature similarity between the video stream features and the preset production finishing features does not meet a preset threshold value, continuously identifying the risk type to which the video stream features belong through a risk identification model.
And S53, rendering all the page components to generate a work risk report.
In the embodiment of the invention, if the feature similarity between the video stream feature and the preset production end feature meets the preset threshold value, the corresponding page components are respectively loaded by adopting all the existing risk information, and all the existing risk information is loaded into the corresponding page components for rendering, so as to generate the working risk report.
As shown in fig. 4, the preset risk recognition model is loaded and trained, that is, the risk recognition model is trained in advance, and meets the preset training standard, and can recognize five preset risk characteristics, wherein the five risk characteristics include a rainy day operation risk characteristic, a non-safety shielding risk characteristic, a non-wearing safety helmet risk characteristic, a construction lighting deficiency risk characteristic in a cable tunnel, and a spray gun illegal use risk characteristic. Inputting a real-time video stream manufactured by a middle head of a heat-shrinkable cable into a risk identification model, extracting video stream characteristics of the real-time video stream, judging whether work is started or not, and continuously judging whether work is started or not when the video stream characteristics do not meet preset manufacturing starting conditions; and when the video stream characteristics meet preset manufacturing starting conditions, identifying the risk type to which the video stream characteristics belong through a risk identification model, wherein the starting characteristics corresponding to the manufacturing starting conditions are that a cable protective layer starts to be stripped and the cable is exposed out of the internal structure. If the risk identification model identifies the risk type to which the video stream features belong, the risk count corresponding to the risk features is +1, where the risk count is equivalent to the number of the above-mentioned constructions. Otherwise the risk count for that risk feature is zeroed. Judging whether the risk count of the risk features is greater than or equal to a preset risk threshold, if the risk count corresponding to any risk feature is greater than or equal to the corresponding preset risk threshold, sending a corresponding key object risk warning, returning the risk count of the risk feature to zero, and then judging whether the work is finished; if the risk count corresponding to any risk feature is smaller than the corresponding preset risk threshold, judging whether the work is finished, namely the video stream feature meets a preset manufacturing finishing condition, wherein the finishing feature corresponding to the manufacturing finishing condition is as follows: the outer sheath pipe is arranged on the interfaces of the two cables, and the outer sheath pipe and the cables are combined under the condition of full thermal shrinkage and tightening. When the video stream characteristics meet the preset manufacturing completion conditions, generating a work risk report of the work; and when the video stream characteristics do not meet the preset manufacturing completion conditions, generating a work risk report of the work, and continuously identifying the risk type to which the video stream characteristics belong by using the risk identification model.
In the embodiment of the invention, firstly, monitoring videos made by a plurality of cable middle heads are obtained, each monitoring video is divided into a plurality of static pictures according to preset division intervals to obtain corresponding static picture sets, risk pictures which are in accordance with preset initial risk types in the static picture sets are extracted, a risk training set and a risk verification set are constructed, an initial risk identification model is trained by adopting the risk training set, and the risk identification model is constructed by combining the risk verification set and preset training standards. Secondly, acquiring real-time video stream between cable middle head production, extracting video stream characteristics of the real-time video stream through a preset risk identification model, identifying risk types to which the video stream characteristics belong through the risk identification model when the video stream characteristics meet preset production starting conditions, constructing risk information by adopting the risk types, counting the constructed quantity of the risk information in real time, and sending risk warning and resetting the constructed quantity when the constructed quantity is larger than or equal to a preset risk threshold value. And finally, when the video stream characteristics meet the preset manufacturing completion conditions, generating a corresponding work risk report by adopting all the existing risk information. The manufacturing process of the cable middle head is monitored in real time through the risk identification model, the risk identification model is adopted to identify the risk characteristics, the identification accuracy can be ensured, a risk warning can be timely sent out when the manufacturing risk occurs, and the safety of the manufacturing process of the cable middle head is guaranteed.
Referring to fig. 5, fig. 5 is a block diagram of a system for identifying a manufacturing risk of a cable middle head according to a third embodiment of the present invention.
The embodiment of the invention provides a system for identifying manufacturing risks of a cable middle head, which comprises:
a real-time video stream acquiring module 501, configured to acquire a real-time video stream between cable middle header productions;
a video stream feature extraction module 502, configured to extract video stream features of a real-time video stream through a preset risk identification model;
a risk type identification module 503, configured to identify, when the video stream characteristics meet a preset production start condition, a risk type to which the video stream characteristics belong through a risk identification model;
a risk information construction and construction quantity statistics module 504, configured to construct risk information by using risk types, and calculate the construction quantity of the risk information in real time;
a risk warning sending and construction number resetting module 505 for sending a risk warning and resetting the construction number when the construction number is greater than or equal to a preset risk threshold;
and a work risk report generating module 506, configured to generate a corresponding work risk report by using existing risk information when the video stream characteristics meet a preset production completion condition.
Optionally, the system further comprises:
and the monitoring video acquisition module is used for acquiring the monitoring videos made by the cable intermediate heads.
And the static picture set building module is used for dividing each monitoring video into a plurality of static pictures respectively according to preset division intervals to obtain corresponding static picture sets.
And the risk training set and risk verification set building module is used for extracting the risk pictures which accord with the preset initial risk types in the static picture set and building a risk training set and a risk verification set.
And the risk identification model building module is used for training the initial risk identification model by adopting a risk training set and building the risk identification model by combining a risk verification set and a preset training standard.
Optionally, the risk identification model building module comprises:
and the target risk training set obtaining module is used for amplifying the number of the risk pictures in the risk training set according to a preset picture amplification mode to obtain a target risk training set.
And the intermediate risk identification model obtaining module is used for training the initial risk identification model by adopting a target risk training set to obtain an intermediate risk identification model.
And the verification structure obtaining module is used for verifying the intermediate risk identification model by adopting a risk verification set to obtain a verification result.
And the risk identification model construction submodule is used for determining the intermediate risk identification model as the risk identification model if the verification result meets the preset training standard.
And the verification result does not meet the preset training standard module, and the step of skipping to execute the step of training the initial risk identification model by adopting the target risk training set to obtain the intermediate risk identification model if the verification result does not meet the preset training standard.
Optionally, the risk identification model includes convolutional layers, and the video stream feature extraction module 502 includes:
and the video frame picture extraction module is used for extracting the video frame pictures of the real-time video stream according to a preset video extraction interval.
And the video stream feature extraction submodule is used for extracting the video stream features of all the video frame pictures through the convolution layer.
Optionally, the risk identification module includes a risk comparison module, and the risk type identification module 503 includes:
and the risk similarity calculation module is used for calculating the risk similarity between the video stream characteristics and the plurality of preset risk characteristics through the risk comparison module when the video stream characteristics meet the preset making starting conditions.
And the risk type selection module is used for selecting the risk type to which the risk characteristic corresponding to the maximum risk similarity belongs as the risk type to which the video stream characteristic belongs.
Optionally, the risk information construction and construction quantity statistics module 504 includes:
and the risk keyword acquisition module is used for acquiring the risk keywords carried by the risk types.
And the risk information extraction and construction information module is used for extracting the safety risk points, the risk occurrence time and the risk pictures corresponding to the risk keywords from a preset video stream characteristic information base.
And the risk information construction and construction quantity counting submodule is used for constructing the risk information by adopting the safety risk points, the risk occurrence time and the risk pictures and counting the construction quantity of the risk information in real time.
Optionally, the risk identification model includes an ending feature comparison module, and the operational risk report generation module 506 includes:
and the characteristic similarity calculation module is used for calculating the characteristic similarity between the video stream characteristic and the preset production ending characteristic through the ending characteristic comparison module.
And the page component loading module is used for respectively loading the corresponding page components by adopting all the existing risk information when the feature similarity meets a preset threshold value.
And the work risk report generation submodule is used for rendering all the page components and generating a work risk report.
An embodiment of the present invention further provides an electronic device, where the electronic device includes: the computer system comprises a memory and a processor, wherein a computer program is stored in the memory; the computer program, when executed by the processor, causes the processor to perform the method for identifying a risk of cable intermediate head manufacturing as in any one of the embodiments described above.
The memory may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. The memory has a memory space for program code for performing any of the method steps of the above-described method. For example, the memory space for the program code may comprise respective program codes for implementing the respective steps in the above method, respectively. The program code can be read from and written to one or more computer program products. These computer program products comprise a program code carrier such as a hard disk, a Compact Disc (CD), a memory card or a floppy disk. The program code may be compressed, for example, in a suitable form. These codes, when executed by a computing processing device, cause the latter to carry out the various steps of the method for identifying risks of cable intermediate head fabrication described above.
The embodiment of the invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for identifying the manufacturing risk of the cable middle head as in any one of the above embodiments is implemented.
It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is only a logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (9)
1. A method for identifying the manufacturing risk of a cable intermediate head is characterized by comprising the following steps:
acquiring a real-time video stream between cable middle head production;
extracting video stream characteristics of the real-time video stream through a preset risk identification model;
when the video stream characteristics meet preset manufacturing starting conditions, identifying the risk type to which the video stream characteristics belong through the risk identification model, wherein the preset manufacturing starting conditions include the picture characteristics that a cable sheath starts to be stripped and a cable exposes out of an internal structure;
adopting the risk types to construct risk information, and counting the constructed quantity of the risk information in real time, wherein the constructed quantity refers to the times of respectively identifying each risk type in the risk identification process;
when the constructed number is larger than or equal to a preset risk threshold, sending a risk warning and resetting the constructed number, wherein the corresponding preset risk threshold is set for each risk type;
when the construction quantity corresponding to any risk type is larger than or equal to the corresponding preset risk threshold value, sending a risk warning and resetting the corresponding construction quantity, and judging whether the video stream characteristics meet preset manufacturing completion conditions or not;
when the construction quantity corresponding to all risk types is smaller than the corresponding preset risk threshold value, judging whether the video stream characteristics meet preset manufacturing completion conditions or not;
when the video stream characteristics meet preset manufacturing completion conditions, generating a corresponding working risk report by adopting all the existing risk information, wherein the preset manufacturing completion conditions refer to picture characteristics that an outer sheath pipe is arranged on the interfaces of two cables, and the outer sheath pipe and the cables are fully shrunk by heat shrinkage;
the steps of constructing risk information by adopting the risk types and counting the constructed quantity of the risk information in real time comprise the following steps:
acquiring a risk keyword carried by the risk type;
extracting a safety risk point, risk occurrence time and a risk picture corresponding to the risk keyword from a preset video stream characteristic information base;
and constructing risk information by adopting the safety risk points, the risk occurrence time and the risk pictures, and counting the constructed quantity of the risk information in real time.
2. The method for identifying a risk of making a cable intermediate head as recited in claim 1, further comprising:
acquiring monitoring videos made by a plurality of cable intermediate heads;
dividing each monitoring video into a plurality of static pictures according to a preset division interval to obtain a corresponding static picture set;
extracting risk pictures which accord with a preset initial risk type in the static picture set, and constructing a risk training set and a risk verification set;
and training an initial risk identification model by adopting the risk training set, and constructing the risk identification model by combining the risk verification set and a preset training standard.
3. The method for identifying the manufacturing risk of the cable intermediate head according to claim 2, wherein the step of training an initial risk identification model by using the risk training set and constructing the risk identification model by combining the risk verification set and a preset training standard comprises:
amplifying the number of the risk pictures in the risk training set according to a preset picture amplification mode to obtain a target risk training set;
training an initial risk identification model by adopting the target risk training set to obtain an intermediate risk identification model;
verifying the intermediate risk identification model by adopting the risk verification set to obtain a verification result;
if the verification result meets a preset training standard, determining the intermediate risk identification model as the risk identification model;
and if the verification result does not meet the preset training standard, skipping to execute the step of training the initial risk identification model by adopting the target risk training set to obtain an intermediate risk identification model.
4. The method for identifying the manufacturing risk of the cable intermediate head as claimed in claim 1, wherein the risk identification model comprises a convolutional layer; the step of extracting the video stream features of the real-time video stream through a preset risk identification model comprises the following steps:
extracting video frame pictures of the real-time video stream according to a preset video extraction interval;
and extracting the video stream characteristics of each video frame picture through the convolution layer.
5. The method for identifying the production risk of the cable intermediate head as claimed in claim 1, wherein the risk identification model comprises a risk comparison module; when the video stream features meet preset production starting conditions, the step of identifying the risk types to which the video stream features belong through the risk identification model comprises the following steps:
when the video stream characteristics meet preset production starting conditions, risk similarity between the video stream characteristics and a plurality of preset risk characteristics is calculated through the risk comparison module;
and selecting the risk type to which the risk feature corresponding to the maximum risk similarity belongs as the risk type to which the video stream feature belongs.
6. The method for identifying the risk of making the cable intermediate head according to claim 1, wherein the risk identification model comprises an ending feature comparison module; when the video stream characteristics meet preset manufacturing completion conditions, the step of generating the corresponding working risk report by adopting all the existing risk information comprises the following steps:
calculating the feature similarity between the video stream features and preset production ending features through the ending feature comparison module;
when the feature similarity meets a preset threshold, respectively loading corresponding page components by adopting all the existing risk information;
rendering all the page components to generate a work risk report.
7. A system for identifying a risk of making a cable intermediate head, comprising:
the real-time video stream acquisition module is used for acquiring a real-time video stream between cable middle head production;
the video stream feature extraction module is used for extracting video stream features of the real-time video stream through a preset risk identification model;
the risk type identification module is used for identifying the risk type of the video stream characteristics through the risk identification model when the video stream characteristics meet preset manufacturing starting conditions, wherein the preset manufacturing starting conditions include that a cable sheath starts to be stripped and a cable exposes the picture characteristics of an internal structure;
the risk information construction and construction quantity counting module is used for constructing risk information by adopting the risk types and counting the construction quantity of the risk information in real time, wherein the construction quantity refers to the times of respectively identifying each risk type in the risk identification process;
the risk warning sending and construction number resetting module is used for sending risk warnings and resetting the construction number when the construction number is larger than or equal to a preset risk threshold, wherein the corresponding preset risk threshold is respectively set for each risk type; when the construction quantity corresponding to any risk type is larger than or equal to the corresponding preset risk threshold value, sending a risk warning and resetting the corresponding construction quantity, and judging whether the video stream characteristics meet preset manufacturing completion conditions or not; when the construction quantity corresponding to all risk types is smaller than the corresponding preset risk threshold value, judging whether the video stream characteristics meet preset manufacturing completion conditions or not;
the working risk report generating module is used for generating a corresponding working risk report by adopting all the existing risk information when the video stream characteristics meet preset manufacturing completion conditions, wherein the preset manufacturing completion conditions include picture characteristics that an outer sheath is installed on interfaces of two cables and the outer sheath and the cables are fully shrunk by heat shrinkage;
the risk information construction and construction quantity statistics module comprises:
a risk keyword obtaining module, configured to obtain a risk keyword carried by the risk type;
the risk information extraction and construction information module is used for extracting a safety risk point, risk occurrence time and a risk picture corresponding to the risk keyword from a preset video stream characteristic information base;
and the risk information construction and construction quantity counting submodule is used for constructing risk information by adopting the safety risk points, the risk occurrence time and the risk pictures and counting the construction quantity of the risk information in real time.
8. An electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, causes the processor to perform the steps of the method for identifying a risk of making a cable middle head according to any one of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed, implements a method of identifying a risk of making a cable intermediate head according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210818638.0A CN114913172B (en) | 2022-07-13 | 2022-07-13 | Method, system, equipment and medium for identifying manufacturing risk of cable middle head |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210818638.0A CN114913172B (en) | 2022-07-13 | 2022-07-13 | Method, system, equipment and medium for identifying manufacturing risk of cable middle head |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114913172A CN114913172A (en) | 2022-08-16 |
| CN114913172B true CN114913172B (en) | 2022-12-30 |
Family
ID=82772342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210818638.0A Active CN114913172B (en) | 2022-07-13 | 2022-07-13 | Method, system, equipment and medium for identifying manufacturing risk of cable middle head |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114913172B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111093054A (en) * | 2019-07-25 | 2020-05-01 | 中国石油天然气股份有限公司 | Intelligent supervision method and system for construction site |
| CN111918039A (en) * | 2020-08-13 | 2020-11-10 | 智点恒创(苏州)智能科技有限公司 | Artificial intelligence high risk operation management and control system based on 5G network |
| CN112184678A (en) * | 2020-09-30 | 2021-01-05 | 国网北京市电力公司 | Image recognition method, image recognition device, computer-readable storage medium and processor |
| CN112257545A (en) * | 2020-10-19 | 2021-01-22 | 安徽领云物联科技有限公司 | Violation real-time monitoring and analyzing method and device and storage medium |
| CN113111728A (en) * | 2021-03-22 | 2021-07-13 | 广西电网有限责任公司电力科学研究院 | Intelligent identification method and system for power production operation risk in transformer substation |
| CN114113105A (en) * | 2021-11-12 | 2022-03-01 | 宁波新胜中压电器有限公司 | Detection equipment and detection method for cable connector |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8140277B2 (en) * | 2009-03-31 | 2012-03-20 | Oracle America, Inc. | Enhanced characterization of electrical connection degradation |
| CN113766256A (en) * | 2021-02-09 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Live broadcast wind control method and device |
| CN113568887A (en) * | 2021-07-30 | 2021-10-29 | 中国工商银行股份有限公司 | Operation and maintenance operation monitoring method and device based on big data platform |
-
2022
- 2022-07-13 CN CN202210818638.0A patent/CN114913172B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111093054A (en) * | 2019-07-25 | 2020-05-01 | 中国石油天然气股份有限公司 | Intelligent supervision method and system for construction site |
| CN111918039A (en) * | 2020-08-13 | 2020-11-10 | 智点恒创(苏州)智能科技有限公司 | Artificial intelligence high risk operation management and control system based on 5G network |
| CN112184678A (en) * | 2020-09-30 | 2021-01-05 | 国网北京市电力公司 | Image recognition method, image recognition device, computer-readable storage medium and processor |
| CN112257545A (en) * | 2020-10-19 | 2021-01-22 | 安徽领云物联科技有限公司 | Violation real-time monitoring and analyzing method and device and storage medium |
| CN113111728A (en) * | 2021-03-22 | 2021-07-13 | 广西电网有限责任公司电力科学研究院 | Intelligent identification method and system for power production operation risk in transformer substation |
| CN114113105A (en) * | 2021-11-12 | 2022-03-01 | 宁波新胜中压电器有限公司 | Detection equipment and detection method for cable connector |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114913172A (en) | 2022-08-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110225299B (en) | Video monitoring method and device, computer equipment and storage medium | |
| CN109271961B (en) | Behavior pattern judgment method based on figure identity feature recognition | |
| CN110674790B (en) | Abnormal scene processing method and system in video monitoring | |
| CN109271881A (en) | Personnel safety management-control method, device and server in a kind of substation | |
| CN113191699A (en) | Power distribution construction site safety supervision method | |
| CN104239309A (en) | Video analysis retrieval service side, system and method | |
| CN209402635U (en) | A kind of monitoring of transformer substation video and intelligent analysis system | |
| CN115953566B (en) | Feature analysis method, system and medium for infrared thermal imager | |
| CN102625083B (en) | Method for constructing video monitoring network | |
| CN112036327A (en) | SSD-based lightweight safety helmet detection method | |
| CN111178241A (en) | Intelligent monitoring system and method based on video analysis | |
| CN111027520A (en) | Method, device and equipment for judging and identifying mountain fire points | |
| CN111860187A (en) | High-precision worn mask identification method and system | |
| CN113743725B (en) | Ascending qualification detection method and system for ascending operators based on edge calculation | |
| CN115567690A (en) | Intelligent monitoring system capable of automatically identifying dangerous points of field operation | |
| CN105407331A (en) | Monitoring method and monitoring equipment | |
| CN114913172B (en) | Method, system, equipment and medium for identifying manufacturing risk of cable middle head | |
| CN116894613B (en) | Visual management system and method applied to airport | |
| CN112231038B (en) | Work order information display method, device, computer equipment and storage medium | |
| CN113627321A (en) | Image identification method and device based on artificial intelligence and computer equipment | |
| CN111553997A (en) | Post changing system and method thereof | |
| CN115546892A (en) | Method, device and storage medium for identifying operation violation | |
| CN115391596A (en) | Video archive generation method and device and storage medium | |
| CN114997582A (en) | Early warning method and system for distribution network operation risk and computer equipment | |
| CN114359828A (en) | Target behavior recording method, device, storage medium and electronic device |
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 |