CN113989711A - Power distribution construction safety tool use identification method and system - Google Patents
Power distribution construction safety tool use identification method and system Download PDFInfo
- Publication number
- CN113989711A CN113989711A CN202111258031.3A CN202111258031A CN113989711A CN 113989711 A CN113989711 A CN 113989711A CN 202111258031 A CN202111258031 A CN 202111258031A CN 113989711 A CN113989711 A CN 113989711A
- Authority
- CN
- China
- Prior art keywords
- image
- coordinate information
- operating rod
- worker
- hand
- 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.)
- Granted
Links
- 238000010276 construction Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000001514 detection method Methods 0.000 claims abstract description 64
- 238000012544 monitoring process Methods 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims description 13
- 238000001914 filtration Methods 0.000 claims description 5
- 238000007689 inspection Methods 0.000 abstract description 25
- 239000000284 extract Substances 0.000 abstract 1
- 238000004519 manufacturing process Methods 0.000 description 20
- 230000006870 function Effects 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000013135 deep learning Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000001681 protective effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
- G06T9/002—Image coding using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a power distribution construction safety tool use identification method and system, and belongs to the technical field of computer vision identification. The invention monitors the electric power operation site in real time, extracts the hand image information, the operating rod image information and the insulating glove image information of the operator from the monitoring video by utilizing a pre-trained target detection model, and further judges the wearing condition of the operator on the operating rod and the insulating glove according to the hand image information, the operating rod image information, the hand image information and the insulating glove image information. The invention can continuously detect the tool service condition of the operating personnel by monitoring the power operation site in real time and analyzing the monitoring video. The problems of high labor intensity and incapability of real-time inspection caused by manual inspection are avoided.
Description
Technical Field
The invention belongs to the technical field of computer vision identification, and particularly relates to a power distribution construction safety tool use identification method and system.
Background
With the increasing demand of production and life on electric power energy, China has generated higher demand on electric power production. Under the influence of various factors, the frequency of electric power production accidents in China is at a higher level, and the urban safety production is threatened. Therefore, an all-round safety production concept must be introduced into the power production operation to guarantee the normal operation of a power system, the process is complex in the power production operation process, high-voltage power equipment can be contacted in the daily inspection and maintenance process, and safety accidents are easy to occur if protective equipment is not properly operated or worn.
In summary, in order to ensure that electric power workers adopt compliance protection measures and operation procedures, the safety protection consciousness of the workers is improved, and the field safety operation level is improved, the traditional mode is that manual inspection is adopted or a supervisor is stationed at each operation field, so that the labor intensity of the inspection personnel is high, the production cost is high, the distance of each operation field is long, most of time of the inspection personnel is wasted on the way, the operation personnel temporarily meet the requirements of the inspection personnel, and after the inspection personnel leave, the operation personnel greet conveniently without adopting compliance protection measures. In recent years, with the gradual maturity of computer vision and internet of things technologies, especially the rapid development of neural network technologies, deep learning technologies are beginning to be applied to various production environments. The current mainstream research direction is to adopt the deep learning related technology to carry out real-time compliance protection measure detection on a construction site.
Disclosure of Invention
In view of the above, the present invention aims to solve the problems of high labor intensity and incapability of real-time inspection in the conventional manual inspection of power production operation.
In order to solve the technical problems, the invention provides the following technical scheme:
in a first aspect, the invention provides a power distribution construction safety tool use identification method, which comprises the following steps:
acquiring a working video image of the electric power working site through a camera rtsp, and decoding;
preprocessing the decoded job video image, and inputting the preprocessed job video image into a pre-trained target detection model to obtain the category, rectangular frame coordinates and scores of a target output by the target detection model;
filtering target information output by the target detection model according to the set category and the set score threshold value to obtain rectangular frame coordinate information of workers, operating rods and insulating gloves;
intercepting a worker image area according to rectangular frame coordinate information of workers, and acquiring hand coordinate information of the workers from the worker image area by using a human body key point detection algorithm;
intercepting an operating rod image area according to rectangular frame coordinate information of the operating rod, detecting the outline of the operating rod, and acquiring outline coordinate information of the operating rod;
whether the worker wears the insulating gloves and whether the worker holds the operating rod or not is judged according to the hand coordinate information of the worker, the rectangular frame coordinate information of the insulating gloves, the hand coordinate information of the worker and the contour coordinate information of the operating rod.
Further, the preprocessing the decoded job video image specifically includes:
and performing quality detection on the decoded working video image, and eliminating black screen, fuzzy, overexposure and jittering pictures in the working video image.
Further, inputting the preprocessed job video image into a pre-trained target detection model to obtain the category, rectangular frame coordinates and score of the target output by the target detection model specifically comprises:
and inputting the preprocessed job video image into a target detection model trained on the basis of a YOLO-v4 algorithm to obtain the category, rectangular frame coordinates and scores of the target output by the target detection model.
Further, judge whether the staff wears insulating gloves according to staff's hand coordinate information and insulating gloves's rectangle frame coordinate information and specifically include:
acquiring a hand area image and an insulating glove area image according to hand coordinate information of workers and rectangular frame coordinate information of insulating gloves;
calculating the intersection of the hand region image and the insulating glove region image by using the image bitwise and the function, and counting the number of white pixels in the intersection;
and judging whether the worker wears the insulating gloves or not according to the ratio of the number of the white pixel points to the total number of the pixels in the intersection.
Further, whether the staff holds the operating rod by hand specifically includes according to the hand coordinate information of staff and the contour coordinate information of operating rod:
acquiring a hand area image and an operating rod outline image according to hand coordinate information of a worker and outline coordinate information of an operating rod;
calculating the intersection of the hand region image and the operating rod outline image by using the image bitwise and the function, and counting the number of white pixels in the intersection;
and judging whether the operator holds the operating rod or not according to the ratio of the number of the white pixels to the total number of the pixels in the intersection.
In a second aspect, the present invention provides a power distribution construction safety tool use identification system, including:
the real-time monitoring unit is used for acquiring an operation video image of the electric power operation site through the camera rtsp and decoding the image;
the target detection unit is used for preprocessing the decoded job video image and inputting the preprocessed job video image into a pre-trained target detection model so as to obtain the category, rectangular frame coordinates and scores of the target output by the target detection model;
the target identification unit is used for filtering target information output by the target detection model according to the set category and the set score threshold value to obtain rectangular frame coordinate information of workers, the operating rod and the insulating gloves;
the hand recognition unit is used for intercepting an image area of the worker according to the rectangular frame coordinate information of the worker and acquiring the hand coordinate information of the worker from the image area of the worker by using a human body key point detection algorithm;
the operating rod identification unit is used for intercepting an operating rod image area according to the rectangular frame coordinate information of the operating rod, detecting the outline of the operating rod and acquiring the outline coordinate information of the operating rod;
the tool uses the recognition unit, and is used for judging whether the worker wears the insulating gloves and whether the worker holds the operating rod or not according to the hand coordinate information of the worker, the rectangular frame coordinate information of the insulating gloves, the hand coordinate information of the worker and the contour coordinate information of the operating rod.
Further, the object detection unit includes: an image preprocessing unit;
the image preprocessing unit is used for detecting the quality of the decoded job video image and eliminating black screen, fuzzy, overexposure and jittering pictures in the job video image.
Further, the object detection unit further includes: an image recognition unit;
the image recognition unit is used for inputting the preprocessed job video image into a target detection model trained on the basis of a YOLO-v4 algorithm so as to obtain the category, rectangular frame coordinates and scores of the target output by the target detection model.
Further, the tool use identification unit includes: an insulating glove use identification unit;
the insulating glove use identification unit is used for acquiring a hand area image and an insulating glove area image according to hand coordinate information of workers and rectangular frame coordinate information of the insulating gloves;
calculating the intersection of the hand region image and the insulating glove region image by using the image bitwise and the function, and counting the number of white pixels in the intersection;
and judging whether the worker wears the insulating gloves or not according to the ratio of the number of the white pixel points to the total number of the pixels in the intersection.
Further, the tool use identification unit includes: an operation lever use identification unit;
the operating rod use identification unit is used for acquiring a hand area image and an operating rod outline image according to hand coordinate information of a worker and outline coordinate information of the operating rod;
calculating the intersection of the hand region image and the operating rod outline image by using the image bitwise and the function, and counting the number of white pixels in the intersection;
and judging whether the operator holds the operating rod or not according to the ratio of the number of the white pixels to the total number of the pixels in the intersection.
In summary, the invention provides a power distribution construction safety tool use identification method and system, which are used for monitoring an electric power working site in real time, extracting hand image information, operating rod image information and insulating glove image information of an operator from a monitoring video by using a pre-trained target detection model, and judging the wearing condition of the operator on an operating rod and an insulating glove according to the hand image information and the operating rod image information and the hand image information and the insulating glove image information respectively. The invention can continuously detect the tool service condition of the operating personnel by monitoring the power operation site in real time and analyzing the monitoring video. The problems of high labor intensity and incapability of real-time inspection caused by manual inspection are avoided.
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 schematic flow chart of a power distribution construction safety tool use identification method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of training a target detection model according to an embodiment of the present invention;
fig. 3 is a schematic view of wearing detection of the operating rod and the insulating gloves according to the embodiment of the invention.
Detailed Description
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, and it is apparent that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
With the increasing demand of production and life on electric power energy, China has generated higher demand on electric power production. Under the influence of various factors, the frequency of electric power production accidents in China is at a higher level, and the urban safety production is threatened. Therefore, an all-round safety production concept must be introduced into the power production operation to guarantee the normal operation of a power system, the process is complex in the power production operation process, high-voltage power equipment can be contacted in the daily inspection and maintenance process, and safety accidents are easy to occur if protective equipment is not properly operated or worn.
In summary, in order to ensure that electric power workers adopt compliance protection measures and operation procedures, the safety protection consciousness of the workers is improved, and the field safety operation level is improved, the traditional mode is that manual inspection is adopted or a supervisor is stationed at each operation field, so that the labor intensity of the inspection personnel is high, the production cost is high, the distance of each operation field is long, most of time of the inspection personnel is wasted on the way, the operation personnel temporarily meet the requirements of the inspection personnel, and after the inspection personnel leave, the operation personnel greet conveniently without adopting compliance protection measures. In recent years, with the gradual maturity of computer vision and internet of things technologies, especially the rapid development of neural network technologies, deep learning technologies are beginning to be applied to various production environments. The current mainstream research direction is to adopt the deep learning related technology to carry out real-time compliance protection measure detection on a construction site.
Based on the above, the embodiment of the invention provides a power distribution construction safety tool use identification method and system, which are used for solving the problems that the labor intensity is high and real-time routing inspection cannot be performed in the conventional manual routing inspection of power production operation.
The following is a detailed description of an embodiment of a power distribution construction safety tool use identification method according to the present invention.
Referring to fig. 1, the present embodiment provides a method for identifying a power distribution construction safety tool, including:
s100: and acquiring a work video image of the electric power work site through the camera rtsp, and decoding.
It should be noted that rtsp (real Time Streaming protocol) is a real-Time Streaming protocol.
And arranging a camera on the electric power operation field to monitor the dead-angle-free coverage of the operation field. And then, acquiring an operation video image in real time through the camera rstp, thereby realizing real-time inspection of the operator.
S200: and preprocessing the decoded job video image, and inputting the preprocessed job video image into a pre-trained target detection model to obtain the category, rectangular frame coordinates and scores of the target output by the target detection model.
It should be noted that, the decoded image is preprocessed by opencv, and image quality detection is performed to remove images with black screen, blur, overexposure and shake, so as to obtain an image with qualified quality; inputting the images with qualified quality into a pre-trained target detection model, wherein the pre-trained target detection model can output the category, rectangular frame coordinates and scores of the target; the required targets here are: the operating personnel, the action bars and insulating gloves.
The target detection model in this embodiment may be a target detection model trained based on the YOLO-v4 algorithm. The training process is shown in fig. 2, namely, the field operation picture is collected; marking the field operation picture, including marking an operator, an operating rod and an insulating glove; then, carrying out data enhancement processing on the image, wherein the data enhancement processing comprises the processing of rotation, mirror image, noise point addition and the like; dividing the processed pictures into a training set and a test set; and modifying the parameters of the YOLO-v4 network for training to finally obtain a target detection model based on YOLO-v 4.
S300: and filtering the target information output by the target detection model according to the set category and the set score threshold value to obtain the rectangular frame coordinate information of the workers, the operating rods and the insulating gloves.
It should be noted that the judgment is performed according to the set score threshold and the category, the target information returned by the target detection model is filtered, some suspected targets with scores lower than the score threshold are removed, and finally the rectangular frame coordinate information of the categories of the operators, the operation rods and the insulating gloves is obtained.
S400: and intercepting an image area of the worker according to the rectangular frame coordinate information of the worker, and acquiring the hand coordinate information of the worker from the image area of the worker by using a human body key point detection algorithm.
It should be noted that, after obtaining the rectangular frame coordinate information of the operator, it is necessary to intercept the image area of the operator, and call the human body key point detection algorithm to obtain the hand coordinate information of the operator, where the obtained hand coordinate is relative to the local image area of the operator, and it is necessary to convert the hand coordinate relative to the entire image.
S400: and intercepting an image area of the operating rod according to the rectangular frame coordinate information of the operating rod, detecting the contour of the operating rod, and acquiring the contour coordinate information of the operating rod.
It should be noted that, after obtaining the coordinate information of the operation lever, the operation lever image area is cut out according to the coordinate information. Because some operating levers are longer, when the operating levers are not in a use state, the image has limitation of a two-dimensional visual angle, and the operating levers can cross the human body area, so that the operating levers can be simply considered to be held by the operator; in order to accurately judge whether an operator holds the operating rod, the contour coordinate of the operating rod needs to be extracted, the contour is detected by using the findcontours function of opencv, the contour of the operating rod can be screened out according to the set contour perimeter threshold, and finally the contour coordinate of the operating rod is obtained.
S600: whether the worker wears the insulating gloves and whether the worker holds the operating rod or not is judged according to the hand coordinate information of the worker, the rectangular frame coordinate information of the insulating gloves, the hand coordinate information of the worker and the contour coordinate information of the operating rod.
It should be noted that, as shown in fig. 3, the specific steps of detecting whether to wear the insulating glove are as follows:
drawing the hand image of an operator on a blank image A according to the hand coordinate of the operator, drawing the image of the insulating glove on a blank image D according to the rectangular frame coordinate of the insulating glove, calling a bitwise _ and function of opencv through the image bitwise _ and function, solving the intersection between the image A and the image D to obtain an image E, counting white pixel points of the image E, finally calculating the proportion of the white pixel points to the total pixels of the superposed image, judging that the operator wears the insulating glove if the proportion exceeds a set threshold value, otherwise, judging that the operator does not wear the insulating glove and needs to send alarm information.
The specific steps of detecting whether the operating rod is held by hand are as follows:
drawing the hand image of an operator in a blank image A according to the hand coordinate of the operator, drawing the operating rod image in a blank image B according to the contour coordinate of the operating rod, calling a bitwise _ and function of opencv through the image bitwise _ and function to obtain the intersection of the image A and the image B, obtaining an image C, counting white pixel points of the image C, calculating the proportion of the white pixel points to the total pixels of the image, judging that the operator holds the operating rod if the proportion exceeds a set threshold value, and otherwise, judging that the operator does not hold the operating rod and needs to send alarm information.
The embodiment provides a power distribution construction safety tool use identification method, which comprises the steps of monitoring an electric power operation site in real time, extracting hand image information, operating rod image information and insulating glove image information of an operator from a monitoring video by using a pre-trained target detection model, and judging the wearing condition of the operator on an operating rod and an insulating glove according to the hand image information and the operating rod image information and the hand image information and the insulating glove image information. The invention can continuously detect the tool service condition of the operating personnel by monitoring the power operation site in real time and analyzing the monitoring video. The problems of high labor intensity and incapability of real-time inspection caused by manual inspection are avoided.
The above is a detailed description of an embodiment of a power distribution construction safety tool usage identification method of the present invention, and the following is a detailed description of an embodiment of a power distribution construction safety tool usage identification system of the present invention.
The embodiment provides a distribution construction safety tool uses identification system includes: the system comprises a real-time monitoring unit, a target detection unit, an image recognition unit, a target recognition unit, a hand recognition unit, an operating rod recognition unit and a tool use recognition unit.
In this embodiment, the real-time monitoring unit is configured to obtain an operation video image of the power operation site through the camera rtsp, and decode the operation video image;
in this embodiment, the target detection unit is configured to pre-process the decoded job video image, and input the pre-processed job video image into a pre-trained target detection model to obtain the category, the rectangular frame coordinates, and the score of the target output by the target detection model.
It should be noted that the target detection unit further includes: an image recognition unit and an image preprocessing unit.
The image recognition unit is used for inputting the preprocessed job video image into a target detection model trained on the basis of a YOLO-v4 algorithm so as to obtain the category, rectangular frame coordinates and scores of the target output by the target detection model.
The image preprocessing unit is used for detecting the quality of the decoded job video image and eliminating black screen, fuzzy, overexposure and jittering pictures in the job video image.
In this embodiment, the target identification unit is configured to filter target information output by the target detection model according to a set category and a score threshold, so as to obtain rectangular frame coordinate information of a worker, an operation rod, and an insulating glove;
in this embodiment, the hand recognition unit is configured to intercept an image area of a worker according to rectangular frame coordinate information of the worker, and acquire hand coordinate information of the worker from the image area of the worker by using a human body key point detection algorithm;
in this embodiment, the operation rod recognition unit is configured to intercept an operation rod image area according to rectangular frame coordinate information of the operation rod, detect an outline of the operation rod, and acquire outline coordinate information of the operation rod;
in this embodiment, the tool use identification unit is used for judging whether the worker wears the insulating gloves and whether the worker holds the operating rod according to the hand coordinate information of the worker, the rectangular frame coordinate information of the insulating gloves, the hand coordinate information of the worker and the contour coordinate information of the operating rod.
The tool use identification unit includes: an insulating glove use identification unit and an operation lever use identification unit.
The insulating glove use identification unit is used for acquiring a hand area image and an insulating glove area image according to hand coordinate information of a worker and rectangular frame coordinate information of the insulating glove; calculating the intersection of the hand region image and the insulating glove region image by using the image bitwise and the function, and counting the number of white pixels in the intersection; and judging whether the worker wears the insulating gloves or not according to the ratio of the number of the white pixel points to the total number of the pixels in the intersection.
The operating rod use identification unit is used for acquiring a hand area image and an operating rod outline image according to hand coordinate information of a worker and outline coordinate information of the operating rod; calculating the intersection of the hand region image and the operating rod outline image by using the image bitwise and the function, and counting the number of white pixels in the intersection; and judging whether the operator holds the operating rod or not according to the ratio of the number of the white pixels to the total number of the pixels in the intersection.
The embodiment provides a distribution construction safety tool uses identification system, realizes the incessant real time monitoring to the electric power field operation through the real time monitoring unit to still use identification unit to discern the surveillance video through the action bars, judge the in service behavior of operation personnel to safety tool automatically. The manpower is saved, the inspection cost is saved, and the inspection efficiency is improved.
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 will 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 (10)
1. A power distribution construction safety tool use identification method is characterized by comprising the following steps:
acquiring a working video image of the electric power working site through a camera rtsp, and decoding;
preprocessing the decoded job video image, and inputting the preprocessed job video image into a pre-trained target detection model to obtain the category, rectangular frame coordinates and fraction of a target output by the target detection model;
filtering the target information output by the target detection model according to the set category and the set score threshold value to obtain rectangular frame coordinate information of workers, operating rods and insulating gloves;
intercepting a worker image area according to rectangular frame coordinate information of workers, and acquiring hand coordinate information of the workers from the worker image area by using a human body key point detection algorithm;
intercepting an operating rod image area according to rectangular frame coordinate information of the operating rod, detecting the outline of the operating rod, and acquiring outline coordinate information of the operating rod;
and judging whether the worker wears the insulating gloves and whether the worker holds the operating rod or not according to the hand coordinate information of the worker, the rectangular frame coordinate information of the insulating gloves, the hand coordinate information of the worker and the contour coordinate information of the operating rod.
2. The method for identifying the use of the power distribution construction safety tool according to claim 1, wherein the preprocessing the decoded work video image specifically comprises:
and performing quality detection on the decoded job video image, and eliminating black screen, fuzzy, overexposure and jittering pictures in the job video image.
3. The power distribution construction safety tool usage identification method according to claim 1, wherein the pre-processed work video image is input into a pre-trained target detection model to obtain the category, rectangular frame coordinates and score of the target output by the target detection model specifically:
inputting the preprocessed job video image into a target detection model trained on the basis of a YOLO-v4 algorithm to obtain the category, rectangular frame coordinates and scores of targets output by the target detection model.
4. The power distribution construction safety tool use identification method according to claim 1, wherein judging whether the worker wears the insulating gloves according to the hand coordinate information of the worker and the rectangular frame coordinate information of the insulating gloves specifically comprises:
acquiring a hand area image and an insulating glove area image according to the hand coordinate information of the worker and the rectangular frame coordinate information of the insulating gloves;
calculating the intersection of the hand region image and the insulating glove region image by using the image according to a position and function, and counting the number of white pixels in the intersection;
and judging whether the worker wears the insulating gloves or not according to the ratio of the number of the white pixels to the total number of the pixels in the intersection.
5. The power distribution construction safety tool use identification method according to claim 1, wherein the step of judging whether the worker holds the operating rod according to the hand coordinate information of the worker and the contour coordinate information of the operating rod specifically comprises the steps of:
acquiring a hand area image and an operating rod outline image according to the hand coordinate information of the worker and the outline coordinate information of the operating rod;
calculating the intersection of the hand region image and the operating rod outline image by using an image bitwise and function, and counting the number of white pixels in the intersection;
and judging whether the operator holds the operating rod or not according to the ratio of the number of the white pixels to the total number of the pixels in the intersection.
6. A power distribution construction safety tool use identification system, comprising:
the real-time monitoring unit is used for acquiring an operation video image of the electric power operation site through the camera rtsp and decoding the image;
the target detection unit is used for preprocessing the decoded job video image and inputting the preprocessed job video image into a pre-trained target detection model so as to obtain the category, rectangular frame coordinates and scores of the target output by the target detection model;
the target identification unit is used for filtering target information output by the target detection model according to the set category and the set score threshold value to obtain rectangular frame coordinate information of workers, operating rods and insulating gloves;
the hand recognition unit is used for intercepting an image area of a worker according to rectangular frame coordinate information of the worker and acquiring hand coordinate information of the worker from the image area of the worker by using a human body key point detection algorithm;
the operating rod identification unit is used for intercepting an operating rod image area according to the rectangular frame coordinate information of the operating rod, detecting the outline of the operating rod and acquiring the outline coordinate information of the operating rod;
the tool use identification unit is used for judging whether the worker wears the insulating gloves or not and whether the worker holds the operating rod or not according to the hand coordinate information of the worker, the rectangular frame coordinate information of the insulating gloves, the hand coordinate information of the worker and the contour coordinate information of the operating rod.
7. The power distribution construction safety tool usage identification system of claim 6, wherein the object detection unit comprises: an image preprocessing unit;
the image preprocessing unit is used for detecting the quality of the decoded job video image and eliminating black screen, fuzzy, overexposure and jittering pictures in the job video image.
8. The power distribution construction safety tool usage identification system of claim 6, wherein the object detection unit further comprises: an image recognition unit;
the image recognition unit is used for inputting the preprocessed job video image into a target detection model trained on the basis of a YOLO-v4 algorithm so as to obtain the category, rectangular frame coordinates and scores of the target output by the target detection model.
9. The power distribution construction safety tool usage identification system of claim 6, wherein the tool usage identification unit comprises: an insulating glove use identification unit;
the insulating glove use identification unit is used for acquiring a hand area image and an insulating glove area image according to the hand coordinate information of the worker and the rectangular frame coordinate information of the insulating glove;
calculating the intersection of the hand region image and the insulating glove region image by using the image according to a position and function, and counting the number of white pixels in the intersection;
and judging whether the worker wears the insulating gloves or not according to the ratio of the number of the white pixels to the total number of the pixels in the intersection.
10. The power distribution construction safety tool usage identification system of claim 6, wherein the tool usage identification unit comprises: an operation lever use identification unit;
the operating rod use identification unit is used for acquiring a hand area image and an operating rod outline image according to the hand coordinate information of the worker and the outline coordinate information of the operating rod;
calculating the intersection of the hand region image and the operating rod outline image by using an image bitwise and function, and counting the number of white pixels in the intersection;
and judging whether the operator holds the operating rod or not according to the ratio of the number of the white pixels to the total number of the pixels in the intersection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111258031.3A CN113989711B (en) | 2021-10-27 | Power distribution construction safety tool use identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111258031.3A CN113989711B (en) | 2021-10-27 | Power distribution construction safety tool use identification method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113989711A true CN113989711A (en) | 2022-01-28 |
| CN113989711B CN113989711B (en) | 2024-07-05 |
Family
ID=
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114758363A (en) * | 2022-06-16 | 2022-07-15 | 四川金信石信息技术有限公司 | Insulating glove wearing detection method and system based on deep learning |
| CN114971983A (en) * | 2022-05-18 | 2022-08-30 | 国网安徽省电力有限公司安庆供电公司 | Electric power marketing operation field management and control system based on intelligent helmet |
| CN115471874A (en) * | 2022-10-28 | 2022-12-13 | 山东新众通信息科技有限公司 | Construction site dangerous behavior identification method based on monitoring video |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111222478A (en) * | 2020-01-11 | 2020-06-02 | 深圳奥腾光通系统有限公司 | Construction site safety protection detection method and system |
| CN113313186A (en) * | 2021-06-09 | 2021-08-27 | 广东电网有限责任公司 | Method and system for identifying non-standard wearing work clothes |
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111222478A (en) * | 2020-01-11 | 2020-06-02 | 深圳奥腾光通系统有限公司 | Construction site safety protection detection method and system |
| CN113313186A (en) * | 2021-06-09 | 2021-08-27 | 广东电网有限责任公司 | Method and system for identifying non-standard wearing work clothes |
Non-Patent Citations (1)
| Title |
|---|
| 孔英会 等: "基于改进Mask R-CNN模型的电力场景目标检测方法", 科学技术与工程, no. 08, 18 March 2020 (2020-03-18), pages 191 - 199 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114971983A (en) * | 2022-05-18 | 2022-08-30 | 国网安徽省电力有限公司安庆供电公司 | Electric power marketing operation field management and control system based on intelligent helmet |
| CN114971983B (en) * | 2022-05-18 | 2024-06-04 | 国网安徽省电力有限公司安庆供电公司 | Electric power marketing operation scene management and control system based on intelligent helmet |
| CN114758363A (en) * | 2022-06-16 | 2022-07-15 | 四川金信石信息技术有限公司 | Insulating glove wearing detection method and system based on deep learning |
| CN114758363B (en) * | 2022-06-16 | 2022-08-19 | 四川金信石信息技术有限公司 | Insulating glove wearing detection method and system based on deep learning |
| CN115471874A (en) * | 2022-10-28 | 2022-12-13 | 山东新众通信息科技有限公司 | Construction site dangerous behavior identification method based on monitoring video |
| CN115471874B (en) * | 2022-10-28 | 2023-02-07 | 山东新众通信息科技有限公司 | Construction site dangerous behavior identification method based on monitoring video |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111445524B (en) | Scene understanding-based construction site worker unsafe behavior identification method | |
| CN110738135B (en) | Method and system for judging and guiding worker operation step standard visual recognition | |
| CN110929591A (en) | Transformer substation monitoring method and device, computer equipment and storage medium | |
| CN111428617A (en) | Video image-based distribution network violation maintenance behavior identification method and system | |
| CN110493574B (en) | Security monitoring visualization system based on streaming media and AI technology | |
| CN112287823A (en) | Facial mask identification method based on video monitoring | |
| CN109308448A (en) | A method of it prevents from becoming distribution maloperation using image processing techniques | |
| CN115802027B (en) | On-line monitoring equipment operation abnormity detection method based on image processing | |
| CN113111771A (en) | Method for identifying unsafe behaviors of power plant workers | |
| CN108174198B (en) | Video image quality diagnosis analysis detection device and application system | |
| CN113239759A (en) | Power grid operation field violation identification method | |
| CN113239760A (en) | Power grid operation field violation identification system | |
| CN114155492A (en) | High-altitude operation safety belt hanging rope high-hanging low-hanging use identification method and device and electronic equipment | |
| CN113808087A (en) | Defect management and control method and device for surface of steel plate and computer readable storage medium | |
| CN110543809A (en) | Electric power operation site risk supervision intrusion video identification method | |
| CN116665419B (en) | Intelligent fault early warning system and method based on AI analysis in power production operation | |
| CN109873990A (en) | A kind of illegal mining method for early warning in mine based on computer vision | |
| CN113297913A (en) | Method for identifying dressing specification of distribution network field operating personnel | |
| CN113989711B (en) | Power distribution construction safety tool use identification method and system | |
| CN112532927A (en) | Intelligent safety management and control system for construction site | |
| CN113989711A (en) | Power distribution construction safety tool use identification method and system | |
| CN113536842A (en) | Electric power operator safety dressing identification method and device | |
| CN114330477B (en) | Power equipment defect detection system and method based on mixed reality equipment | |
| CN115601712A (en) | Image data processing method and system suitable for field safety measures | |
| WO2022121886A1 (en) | Method and apparatus for identifying dress code for electric power operations |
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 |