CN113297914B - Distribution network field operation electricity testing action recognition method - Google Patents
Distribution network field operation electricity testing action recognition method Download PDFInfo
- Publication number
- CN113297914B CN113297914B CN202110453958.6A CN202110453958A CN113297914B CN 113297914 B CN113297914 B CN 113297914B CN 202110453958 A CN202110453958 A CN 202110453958A CN 113297914 B CN113297914 B CN 113297914B
- Authority
- CN
- China
- Prior art keywords
- video
- electricity testing
- action
- model
- distribution 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
-
- 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
Abstract
The invention provides a distribution network field operation electricity testing action identification method, which comprises the following steps: model construction and training: classifying the video frames, and correspondingly training YOLOv4 and MobileNetv2 models on the classified video frames; video clip generation: the method comprises the steps that targets exist in uploaded distribution network field operation videos, and video clips possibly with electricity testing actions are generated based on detection results; whether an electroscopic action exists: detecting the uploaded distribution network field operation video, respectively carrying out target detection and target classification on the existing electricity testing equipment and/or video frames related to the electricity testing action, and judging whether the electricity testing action exists or not according to the comprehensive judgment; identifying the electroscopy action: and restoring the video frames with the electroscopic actions into videos, and identifying whether the electroscopic actions exist. According to the invention, through extracting the characteristics of the electricity testing action and utilizing the deep learning algorithm to identify the electricity testing action, the workload of video auditors is reduced, the working efficiency is improved and the labor cost is reduced.
Description
Technical Field
The invention relates to the technical field of distribution network operator standard operation, in particular to a distribution network field operation electricity testing action identification method.
Background
The current social economy is rapidly increased, the electricity consumption is increased year by year, the scale of distribution network equipment for providing power supply for users is sharply expanded, the distribution network structure is more and more complex, factors influencing the operation stability of the distribution network are increased, the distribution network faults are frequent, and the operation and maintenance strength is caught. And along with the development of the manufacturing process towards the refinement direction and the improvement of the requirements of various electric equipment on the electric energy quality, the method not only puts forward a new requirement on the power supply reliability, but also generates new cognition on the power supply quality.
From the current requirements and trends of various industries on power supply, the distribution network develops towards the trend of initiative, intellectualization, green, reliability and high efficiency in the future. Therefore, the power supply unit strengthens the power supply reliability, improves the power quality and the power supply capacity, avoids the serious economic loss caused by large-area power failure due to the damage to the distribution network line caused by natural environment, severe weather and external force damage, and is vital to avoid power grid accidents and ensure the line safety. How to improve the operation stability of distribution network equipment, accelerate the timeliness of monitoring and operation and maintenance of a power supply line, strengthen the early prevention of causing line faults, improve the operation efficiency of distribution network rush repair and strengthen the safety management and control of the operation process is urgent. When the electrical equipment needs grounding operation under the power failure operation, electricity must be tested firstly, and after the fact that no voltage exists is proved, the operation of connecting a grounding switch or installing a grounding wire can be carried out. If the preparation work before the electricity test is not carried out according to relevant regulations and the electricity test method is not complied with, personal accidents of workers can be caused. The electricity testing action is checked by a manual mode, and the effect that each video cannot be checked is poor due to overlong and overlarge videos of field operation, so that the electricity testing operation in the field safety operation of intelligent model detection is very necessary. By extracting the characteristics of the electricity testing action and recognizing the electricity testing action by using a deep learning algorithm, the workload of video auditors is reduced, the work efficiency is improved, the labor cost is reduced, and the identification rate of the electricity testing action is improved.
Disclosure of Invention
The invention aims to provide an electroscopic action video identification method for identifying electroscopic action by utilizing a deep learning algorithm.
In order to achieve the above object, the present invention is achieved by the following means.
A distribution network field operation electricity testing action recognition method comprises the following steps:
model construction and training: classifying the video frames, and correspondingly training YOLOv4 and MobileNetv2 models on the classified video frames;
video clip generation: in the uploaded distribution network field operation video, targets related to the electricity testing equipment and/or the electricity testing action exist for detection, and video clips possibly with the electricity testing action are generated based on detection results;
whether an electroscopy action exists: detecting the uploaded distribution network field operation video, respectively carrying out target detection and target classification on the existing electricity testing equipment and/or video frames related to the electricity testing action, and comprehensively judging whether the electricity testing action exists according to target detection and target classification results;
and (3) identifying the electroscopy action: and restoring the video frames with the electroscopic actions into videos, and identifying whether the electroscopic actions exist or not by utilizing the videos.
As a further improvement of the present invention, before the model is built and trained, the method further comprises the following steps of obtaining video frames: and collecting the process video of the distribution network field operation, and intercepting the video with the electricity testing part in the process video to form a video frame.
As a further improvement of the present invention, in the step of acquiring the video frame, the method further includes processing the video frame, where the processing specifically includes: the video frames in the form of pictures are feature scaled so that the pixel values of the graphics pixels lie in the [0,1] interval.
As a further improvement of the present invention, in the step of model construction and training, the classifying the video frames specifically includes: and respectively forming a target detection model and a target classification model by using the video frames which are related to the electricity testing equipment and/or the electricity testing action, and respectively training each model to generate a model1 and a model2.
As a further improvement of the invention, the model construction and training steps are specifically as follows: taking a picture with the action that a handheld electroscope contacts an electric wire in a video frame as a training set of a YOLOv4 model, and training the model to generate a model1; and (3) taking pictures with electroscope and grounding rod in the video frame and other rod-shaped objects as a training set of the MobileNetv2 model, training the model, and generating the model2.
As a further improvement of the present invention, the video segment generating step specifically includes: and (3) carrying out preliminary detection on the uploaded distribution network field operation videos by using a model1, marking the section of the videos if detecting that the electricity testing equipment or the electricity testing action exists, generating video segments with the electricity testing action, and skipping the videos if not.
As a further improvement of the present invention, the step of determining whether there is an electroscopy action specifically comprises: and selecting the marked video segments, analyzing the videos by using a model1 and a model2 respectively, and comprehensively judging whether an electroscopic action exists or not based on the result.
As a further improvement of the present invention, the electroscopic action identification specifically includes: and based on the threshold value, fusing the video frames with the electroscopic action in the detection results of the model1 and the model2, restoring the video frames into a video, and identifying whether the electroscopic action exists according to the video.
The invention has the following beneficial effects:
firstly, detecting and positioning distribution network field operators, identifying work clothes and dividing work clothes masks through Mask-RCNN, and then intercepting the positioned operators from an original image; and classifying and identifying whether the distribution network field operation personnel wear the working clothes normally or not through the MobileNet-v2 to obtain the identification results of whether the distribution network field operation personnel wear the working clothes normally or not, wear the working clothes normally or not and wear the working clothes normally, and labeling and positioning the distribution network operation personnel image with the safety belt by adopting class activation mapping to realize the visualization of the image identification result.
The method replaces the manual spot check of the monitoring center staff for whether the distribution network field operation staff wears the working clothes and the standard wearing of the working clothes, and the service scene test result of the distribution network field operation shows that the accuracy rate of the identification method reaches 98.5 percent, and the method can be applied to the identification of whether the distribution network field operation staff wears the working clothes and the standard wearing of the working clothes.
Drawings
Fig. 1 is a schematic flow chart of a distribution network field operation electricity testing action recognition method provided by the invention.
FIG. 2 is a standard convolutional neural network convolution kernel provided by the present invention.
Fig. 3 is a depth separable convolution as provided by the present invention.
Fig. 4 is a 1 x 1 depth separable convolution as provided by the present invention.
Fig. 5 is a process diagram of the process of the present invention for inverting residual error.
Fig. 6 is a process diagram of residual errors provided by the present invention.
Detailed Description
The present invention is described in detail with reference to the embodiments shown in the drawings, but it should be understood that these embodiments are not intended to limit the present invention, and those skilled in the art should understand that functional, methodological, or structural equivalents or substitutions made by these embodiments are within the scope of the present invention.
Example 1
The technical scheme of the invention comprises the following steps:
referring to fig. 1 to 5, in this embodiment, a distribution network field operation electricity testing action identification method includes the following steps:
the method comprises the following steps of firstly, acquiring a video frame, wherein the video frame acquisition comprises the following steps: collecting a process video of distribution network field operation, and intercepting a video with an electricity testing part in the process video to form a video frame;
further, the method further comprises processing the video frame, wherein the processing specifically comprises: the video frames in the form of pictures are feature scaled so that the pixel values of the graphics pixels lie in the [0,1] interval.
Secondly, model construction and training: classifying the video frames, and correspondingly training YOLOv4 and MobileNetv2 models on the classified video frames;
the method specifically comprises the following steps: and respectively forming a target detection model and a target classification model for the video frames with the related electroscopic equipment and/or electroscopic actions, respectively training each model, and generating a model1 and a model2.
Further, the model construction and training steps are specifically as follows: taking a picture with the action that a handheld electroscope contacts an electric wire in a video frame as a training set of a YOLOv4 model, and training the model to generate a model1; and (3) taking pictures with electroscope and grounding rod in the video frame and other rod-shaped objects as a training set of the MobileNetv2 model, training the model, and generating the model2.
Thirdly, generating a video clip: in the uploaded distribution network field operation video, the existence of an electricity testing device and/or a target related to an electricity testing action is detected, and a video clip possibly having the electricity testing action is generated based on a detection result;
further, the video clip generation step specifically includes: and (3) carrying out preliminary detection on the uploaded distribution network field operation videos by using a model1, marking the section of the videos if detecting that the electricity testing equipment or the electricity testing action exists, and generating a video segment with the electricity testing action, otherwise, skipping the videos.
Step four, whether an electricity testing action exists or not: detecting the uploaded distribution network field operation video, respectively carrying out target detection and target classification on the existing electricity testing equipment and/or video frames related to the electricity testing action, and comprehensively judging whether the electricity testing action exists according to target detection and target classification results;
the method specifically comprises the following steps: and selecting the marked video segments, analyzing the videos by respectively using the model1 and the model2, and comprehensively judging whether the electricity testing action exists or not based on the result.
Fifthly, identifying the electroscopy action: and for the video frames with the electroscopic actions, restoring the video frames into videos, and identifying whether the electroscopic actions exist by utilizing the videos.
The method comprises the following specific steps: and based on the threshold value, fusing the video frames with the electricity testing action in the detection results of the model1 and the model2, restoring the video frames into a video, and identifying whether the electricity testing action exists according to the video.
Example 2
In this embodiment, description is made with reference to specific data.
Referring to fig. 1-5, flow charts of field operation electricity testing action recognition provided by the embodiment of the invention are shown, including:
s101: and (6) data processing.
Collecting videos of electricity testing operation existing in the distribution network field operation process, editing the videos of the electricity testing part, processing the videos into a video frame form, and processing the video frame into pictures with the size of 512 by 512. And because the processed picture is a color image and the stable characteristic does not exist among color channels, the picture needs to be subjected to characteristic scaling so that the pixel value of the image is positioned in the interval of [0,1 ].
S102: constructing a target detection model and constructing a target classification model.
And respectively constructing a target detection model and a target classification model according to the training set data collected and processed in the step S101. The target detection model is a YOLOv4 model, and the target classification model is a MobileNetv2 model. Using a picture of the action that a hand holds an electroscope to contact the wire after data processing is finished as a training set of a YOLOv4 model, training the model and generating a model1; and (3) taking pictures of electroscope and grounding rod and other rod-shaped object pictures in the video frame as a training set of the MobileNetv2 model, training the model, and generating a model2.
(1) Two key points of the MobileNetv2 network are the use of the reciprocal residual structure and the depth separable convolution.
The depth separable convolution is compared with the conventional convolution as shown in FIG. 2
The contrast between the inverted residual and the residual is shown in FIG. 3
S103: detecting a video by using a model1, generating a video segment, after uploading a job video in a field work recorder, firstly grouping and storing the job video by taking a job team as a unit, secondly processing the uploaded video into a video frame format, processing the video frame into a picture with the size of 448 x 448, secondly primarily detecting the picture by using the model1, labeling the video segment if the existence of an electricity testing device or an electricity testing action is detected, generating the video segment with the electricity testing action, and skipping the video if the existence of the electricity testing device or the electricity testing action is not detected;
s104: analysis of the preliminarily detected video clips using model1, model2
Processing the video clips which pass the preliminary detection into a video frame form, processing the video frames into pictures with the size of 448 x 448, respectively analyzing the processed pictures by using a model1 and a model2, and comprehensively judging whether the electricity testing action exists or not based on a model result;
s105: according to the threshold value, obtaining the prompt of whether the electricity testing action exists
And according to the determined threshold value, the detection results of the model1 and the model2 are fused, and the video frame is restored into a video to obtain a prompt of whether the electricity testing action exists in the video.
The invention discloses a distribution network field operation electricity testing action recognition method which realizes recognition of electricity testing actions in a distribution network field operation process through target inspection (YOLOv 4), target tracking (Deepsort), a classification recognition model (MobileNet 2) and an action recognition algorithm. Firstly, objects related to the electroscopic work behaviors, such as electroscopic equipment in a video frame, are detected, and a video clip with possible electroscopic action behaviors is generated based on the detection result. Secondly, respectively carrying out target detection and target classification on the video frames in the video clips, and comprehensively judging whether the video frames have electroscopy actions or not based on the detection and classification results. The target detection model and the target classification model are trained in advance and are deployed in an electroscopic action recognition framework. The target detection model is pre-trained through the electricity testing characteristic that the electroscope is held by a person to contact a wire pole, and the target classification model is used for classifying and training the electroscope, the grounding rod and other rod-shaped objects. And finally, fusing the detection results of the video frames based on the threshold value to obtain a prompt of whether the electricity testing action exists in the video.
According to the invention, through extracting the characteristics of the electricity testing action, the electricity testing action is identified by using a deep learning algorithm, and the electricity testing action identification by relying on a manual mode in the prior operation is replaced, so that the workload of video auditors is reduced, the working efficiency is improved, and the labor cost is reduced.
The steps of the present invention, in brief, comprise the steps of:
1) Carrying out the training of a YOLOv4 and MobileNetv2 model;
2) Detecting targets related to the electricity testing action, such as electricity testing equipment in the video frame, and generating a video clip possibly having the electricity testing action based on the detection result;
3) Respectively carrying out target detection and target classification on video frames in the video clips with the electroscopic actions, and comprehensively judging whether the electroscopic actions exist in the video frames based on the detection and classification results;
4) And based on the threshold value, fusing the detection results of the video frames to obtain a prompt of whether the video has the electricity testing action.
According to the method, the characteristics of the electroscopic action are extracted, and the electroscopic action is identified by using a deep learning algorithm, so that the workload of video auditors is reduced, the working efficiency is improved, the labor cost is reduced, and the electroscopic action identification rate is improved.
The invention discloses a distribution network field operation electricity testing action recognition method which realizes recognition of electricity testing actions in a distribution network field operation process through target detection (YOLOv 4), target tracking (Deepsort), a classification recognition model (MobileNetv 2) and an action recognition algorithm. Firstly, objects related to the electroscopic work behaviors, such as electroscopic equipment in a video frame, are detected, and a video clip with possible electroscopic action behaviors is generated based on the detection result. Secondly, respectively carrying out target detection and target classification on the video frames in the video clips, and comprehensively judging whether the video frames have electroscopy actions or not based on the detection and classification results. The target detection model and the target classification model are trained in advance and are deployed in an electroscopic action recognition framework. The target detection model is pre-trained through the electricity testing characteristic that the electroscope is held by a person to contact a wire pole, and the target classification model is used for classifying and training the electroscope, the grounding rod and other rod-shaped objects. And finally, fusing the detection results of the video frames based on the threshold value to obtain a prompt of whether the electricity testing action exists in the video. According to the invention, through extracting the characteristics of the electricity testing action, the electricity testing action is identified by using a deep learning algorithm, and the electricity testing action identification by relying on a manual mode in the prior operation is replaced, so that the workload of video auditors is reduced, the working efficiency is improved, and the labor cost is reduced.
The above-listed detailed description is only a specific description of a possible embodiment of the present invention, and they are not intended to limit the scope of the present invention, and equivalent embodiments or modifications made without departing from the technical spirit of the present invention should be included in the scope of the present invention.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (7)
1. A distribution network field operation electricity testing action recognition method is characterized by comprising the following steps:
model construction and training: classifying the video frames, and correspondingly training YOLOv4 and MobileNetv2 models on the classified video frames;
the YOLOv4 model and the MobileNetv2 model are respectively a target detection model and a target classification model;
the training of the YOLOv4 and MobileNetv2 models specifically comprises the following steps: after the data processing is finished, a picture of the action that a person holds an electroscope to contact with a wire is used as a training set of a YOLOv4 model, the model is trained, and a model1 is generated; taking pictures of electroscopes and ground rods and pictures of other rod-shaped objects in a video frame as a training set of a MobileNetv2 model, training the model, and generating a model2;
video clip generation: in the uploaded distribution network field operation video, targets related to the electricity testing equipment and/or the electricity testing action exist for detection, and video clips possibly with the electricity testing action are generated based on detection results;
whether an electroscopy action exists: detecting the uploaded distribution network field operation video, respectively carrying out target detection and target classification on the existence of the electricity testing equipment and/or video frames related to the electricity testing action, and comprehensively judging whether the electricity testing action exists according to target detection and target classification results;
identifying the electroscopy action: restoring the video frames with the electroscopic actions into videos, and identifying whether the electroscopic actions exist or not by utilizing the videos;
the electroscopic action identification specifically comprises the following steps: and based on the threshold value, fusing the video frames with the electricity testing action in the detection results of the model1 and the model2, restoring the video frames into a video, and identifying whether the electricity testing action exists according to the video.
2. The distribution network field operation electricity testing action recognition method according to claim 1, characterized in that the step of model building and training further comprises video frame acquisition, wherein the video frame acquisition is as follows: collecting the process video of the power distribution network field operation, and intercepting the video with the electricity testing part in the process video to form a video frame.
3. The distribution network field operation electricity testing action identification method according to claim 2, wherein the video frame acquisition step further comprises processing the video frame, and the processing specifically comprises: the video frames in the form of pictures are feature scaled so that the pixel values of the graphics pixels lie in the [0,1] interval.
4. The distribution network field operation electricity testing action recognition method according to claim 1, wherein in the model building and training step, the video frame classification specifically comprises: and respectively forming a target detection model and a target classification model for the video frames with the related electroscopic equipment and/or electroscopic actions, respectively training each model, and generating a model1 and a model2.
5. The distribution network field operation electricity testing action recognition method according to claim 4, wherein the model construction and training steps specifically include: taking a picture with the action that a handheld electroscope contacts an electric wire in a video frame as a training set of a YOLOv4 model, and training the model to generate a model1; and (3) taking pictures with electroscope and grounding rod in the video frame and other rod-shaped objects as a training set of the MobileNetv2 model, training the model, and generating the model2.
6. The distribution network field operation electricity testing action identification method according to claim 4, wherein the video clip generation step specifically comprises: and (3) carrying out preliminary detection on the uploaded distribution network field operation videos by using a model1, marking the section of the videos if detecting that the electricity testing equipment or the electricity testing action exists, generating video segments with the electricity testing action, and skipping the videos if not.
7. The distribution network field operation electricity testing action identification method according to claim 6, wherein the step of determining whether the electricity testing action exists specifically comprises the steps of: and selecting the marked video segments, analyzing the videos by using a model1 and a model2 respectively, and comprehensively judging whether an electroscopic action exists or not based on the result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110453958.6A CN113297914B (en) | 2021-04-26 | 2021-04-26 | Distribution network field operation electricity testing action recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110453958.6A CN113297914B (en) | 2021-04-26 | 2021-04-26 | Distribution network field operation electricity testing action recognition method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113297914A CN113297914A (en) | 2021-08-24 |
| CN113297914B true CN113297914B (en) | 2023-04-18 |
Family
ID=77320176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110453958.6A Active CN113297914B (en) | 2021-04-26 | 2021-04-26 | Distribution network field operation electricity testing action recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113297914B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112507945A (en) * | 2020-12-18 | 2021-03-16 | 南方电网电力科技股份有限公司 | Method and device for managing and controlling behavior of operator, electronic equipment and storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200348662A1 (en) * | 2016-05-09 | 2020-11-05 | Strong Force Iot Portfolio 2016, Llc | Platform for facilitating development of intelligence in an industrial internet of things system |
| EP3622438A4 (en) * | 2017-05-09 | 2021-03-10 | Neurala, Inc. | Systems and methods to enable continual, memory-bounded learning in artificial intelligence and deep learning continuously operating applications across networked compute edges |
| CN109145789A (en) * | 2018-08-09 | 2019-01-04 | 炜呈智能电力科技（杭州）有限公司 | Power supply system safety work support method and system |
| CN112016461A (en) * | 2020-08-28 | 2020-12-01 | 深圳市信义科技有限公司 | Multi-target behavior identification method and system |
| CN112183313B (en) * | 2020-09-27 | 2022-03-11 | 武汉大学 | SlowFast-based power operation field action identification method |
| CN112347889B (en) * | 2020-10-29 | 2024-04-05 | 广东电网有限责任公司电力科学研究院 | Substation operation behavior identification method and device |
| CN112562439A (en) * | 2020-12-07 | 2021-03-26 | 国家电网有限公司华东分部 | Electric power safety tool real object virtualization device |
-
2021
- 2021-04-26 CN CN202110453958.6A patent/CN113297914B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112507945A (en) * | 2020-12-18 | 2021-03-16 | 南方电网电力科技股份有限公司 | Method and device for managing and controlling behavior of operator, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113297914A (en) | 2021-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108416294B (en) | Fan blade fault intelligent identification method based on deep learning | |
| CN108537154B (en) | Power transmission line bird nest identification method based on HOG characteristics and machine learning | |
| CN103069434B (en) | For the method and system of multi-mode video case index | |
| CN110569841B (en) | Contact gateway key component target detection method based on convolutional neural network | |
| CN107133592B (en) | Human body target feature detection algorithm for power substation by fusing infrared thermal imaging and visible light imaging technologies | |
| CN109060826B (en) | Wind-powered electricity generation blade detection device that does not shut down | |
| CN108680833B (en) | Composite insulator defect detection system based on unmanned aerial vehicle | |
| CN111222478A (en) | Construction site safety protection detection method and system | |
| CN115294117B (en) | Defect detection method and related device for LED lamp beads | |
| CN110458126B (en) | Pantograph state monitoring method and device | |
| CN111476083A (en) | Automatic identification method for wearing of safety helmet of electric power staff | |
| CN113553977A (en) | Improved YOLO V5-based safety helmet detection method and system | |
| CN116231504A (en) | Remote intelligent inspection method, device and system for booster station | |
| CN111951247B (en) | Power equipment ultraviolet image diagnosis method and system | |
| CN113297914B (en) | Distribution network field operation electricity testing action recognition method | |
| CN113297913A (en) | Method for identifying dressing specification of distribution network field operating personnel | |
| CN113066070A (en) | Multi-source data fusion interaction method in three-dimensional scene | |
| CN113160202A (en) | Crack detection method and system | |
| CN116862843A (en) | Construction site safety risk noninductive inspection method and system | |
| KR20210037199A (en) | Apparatus for dividing, tagging an image and for detecting defect of facilities using the same | |
| CN113449767B (en) | Multi-image fusion transformer substation equipment abnormity identification and positioning method | |
| CN112132826A (en) | Pole tower accessory defect inspection image troubleshooting method and system based on artificial intelligence | |
| CN113139476A (en) | Data center-oriented human behavior attribute real-time detection method and system | |
| CN112926488A (en) | Operating personnel violation identification method based on electric power tower structure information | |
| CN112347889A (en) | Substation operation behavior identification method and 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 |