CN112215138A - Deep learning-based violation detection method for low hanging height of safety belt - Google Patents
Deep learning-based violation detection method for low hanging height of safety belt Download PDFInfo
- Publication number
- CN112215138A CN112215138A CN202011082084.XA CN202011082084A CN112215138A CN 112215138 A CN112215138 A CN 112215138A CN 202011082084 A CN202011082084 A CN 202011082084A CN 112215138 A CN112215138 A CN 112215138A
- Authority
- CN
- China
- Prior art keywords
- safety belt
- detection
- hanging
- data
- key point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- 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
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- 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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a safety belt low-hanging high-utilization violation detection method based on deep learning, and provides a novel overhead working safety belt low-hanging high-utilization violation detection algorithm based on a Mask RCNN algorithm aiming at the safety belt specification problem of electric power overhaul workers, so that the problem of operator safety belt hanging position violation detection is completed in real time and high efficiency. Aiming at the problems of complexity, scene variability and the like of the phenomenon of safety belt hanging violation, the invention provides a detection method of a Mask-Keypoints RCNN novel overhead working safety belt violation hanging method which is practically used for safety belt detection and human body key point information combined detection.
Description
Technical Field
The invention relates to machine learning, deep learning computer vision and graphic image processing, in particular to a method for detecting violation of low hanging height of a safety belt based on deep learning.
Background
At present, the hanging position of a safety belt for high-altitude operation electric power overhaul has two modes of low hanging height and high hanging height, and hanging the hanging ring part of the safety belt to a position above the waist of a worker is called the high hanging height of the safety belt, so that the impact distance when an accidental falling accident occurs is favorably reduced, and the hanging method is a safe, normal and scientific hanging method. The other method is low hanging height, the safety belt hanging ring is hung below the waist of the worker, and the safety belt cannot play a role in safety buffering when the worker falls. Therefore, the safety belt must be hung from high to low, and the low hanging from high is avoided.
In the construction site of electric power overhaul high-altitude operation, the overlapped and crossed high-altitude operation is complicated and various. In order to enhance the personal safety of workers and prevent the falling accidents of the workers during the high-altitude operation, enterprise management departments stipulate that the workers must wear safety belts during construction and maintenance according to relevant regulations. However, the problem that the detection of the safety belt area is incomplete due to the complexity of a field scene, namely, the situation that a part of areas are not detected exists, and the judgment difficulty is high for the illegal use of low-hang-height, so that the target detection algorithm is required to be robust enough and can adapt to the complex scene of high-altitude operation. The intelligent safety belt violation detection algorithm is designed to be of great importance, the supervision efficiency of the safety belt wearing condition of high-altitude operation personnel can be greatly improved, the intelligent safety management of the violation personnel can be realized, and the safety of workers is guaranteed.
Disclosure of Invention
In order to solve the defects and shortcomings in the prior art, the invention provides a deep learning-based violation detection method for low hanging height of a safety belt.
The technical scheme of the invention is as follows:
a deep learning-based violation detection method for low hanging height of a safety belt is characterized by comprising the steps of image preprocessing, image enhancement, human body key point detection and safety belt hanging position judgment. The method comprises the following steps:
step (1), video frame data coding buffering, and cutting a video file into an original picture;
preprocessing the data set in the step (1), wherein the mainly adopted data enhancement method comprises contrast change, vertical turning and horizontal turning, random rotation, random cutting and the like;
step (3), marking and training a personnel key point position detection module by using the data in the step (2), and intercepting a personnel information detection boundary frame at a part below a knee of a key point;
step (4), performing data enhancement on the safety belt data cut in the step (3), expanding data quantity, performing safety belt data labeling by using a labeling tool to obtain a corresponding target JSON file, performing feature extraction by using a convolutional neural network, and training a corresponding safety belt detection model;
and (5) combining the safety belt detection module and the human body key point module to judge the hanging type of the safety belt, further judging whether the safety belt is worn normally, and returning a detection result.
The invention has the beneficial effects that:
(1) the deep learning technology and the image processing technology are combined, and Mask-Keypoints RCNN is used for simultaneously realizing safety belt detection and human body key point detection, so that the execution efficiency of the algorithm is improved;
(2) the paper adopts a Mask-Keypoints RCNN efficient algorithm combining a key point positioning technology and safety belt detection, judges whether safety belt hanging position is standard or not according to whether safety belts exist at key points of important bones in the area below the knee or not, can adapt to complex scenes of high-altitude operation, and has robustness and simplicity.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a general flow chart of a deep learning-based violation detection method for low hanging and high hanging of a safety belt in the invention;
FIG. 2 is a diagram of a feature extraction network detection process of the deep learning-based violation detection method for low hanging height of a safety belt;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the method for detecting the violation of hanging high safety belt based on deep learning of the present invention comprises the following basic steps: and preprocessing the acquired image, enhancing the image, detecting a safety belt and a human body key point, and judging the hanging position of the safety belt.
The following describes in detail the violation detection method for low-hanging and high-hanging safety belts based on deep learning with reference to fig. 1 and 2:
step (1), video frame data coding buffering, and cutting a video file into an original picture;
preprocessing the data set in the step (1), wherein the mainly adopted data enhancement method comprises contrast change, vertical turning and horizontal turning, random rotation, random cutting and the like;
step (3), marking and training a personnel key point position detection module by using the data in the step (2), and intercepting a personnel information detection boundary frame at a part below a knee of a key point;
step (4), performing data enhancement on the safety belt data cut in the step (3), expanding data quantity, performing safety belt target marking by using a marking tool to obtain a corresponding target JSON file, performing feature extraction by using a convolutional neural network, and training a corresponding safety belt detection model;
and (5) combining the safety belt detection module and the human body key point module to judge the hanging type of the safety belt, further judging whether the safety belt is worn normally, and returning a detection result.
The invention relates to a safety belt low-hanging high-utilization violation detection method based on deep learning, which combines a deep learning technology with an image processing technology, provides a Mask-Keypoints RCNN novel high-altitude operation safety belt violation hanging method detection method which is practically used for safety belt detection and human body key point information combined detection.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (1)
1. A deep learning-based violation detection method for low hanging height of a safety belt is characterized by comprising the steps of image preprocessing, image enhancement, human body key point detection and safety belt hanging position judgment. The method comprises the following steps:
step (1), video frame data coding buffering, and cutting a video file into an original picture;
preprocessing the data set in the step (1), wherein the mainly adopted data enhancement method comprises contrast change, vertical turning and horizontal turning, random rotation, random cutting and the like;
step (3), marking and training a personnel key point position detection module by using the data in the step (2), and intercepting a personnel information detection boundary frame at a part below a knee of a key point;
step (4), performing data enhancement on the safety belt data cut in the step (3), expanding data quantity, performing safety belt data labeling by using a labeling tool to obtain a corresponding target JSON file, performing feature extraction by using a convolutional neural network, and training a corresponding safety belt detection model;
and (5) combining the safety belt detection module and the human body key point module to judge the hanging type of the safety belt, further judging whether the safety belt is worn normally, and returning a detection result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011082084.XA CN112215138A (en) | 2020-10-12 | 2020-10-12 | Deep learning-based violation detection method for low hanging height of safety belt |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011082084.XA CN112215138A (en) | 2020-10-12 | 2020-10-12 | Deep learning-based violation detection method for low hanging height of safety belt |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112215138A true CN112215138A (en) | 2021-01-12 |
Family
ID=74053222
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011082084.XA Pending CN112215138A (en) | 2020-10-12 | 2020-10-12 | Deep learning-based violation detection method for low hanging height of safety belt |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112215138A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113392708A (en) * | 2021-05-13 | 2021-09-14 | 上海湃道智能科技有限公司 | Safety belt detection method |
CN114155492A (en) * | 2021-12-09 | 2022-03-08 | 华电宁夏灵武发电有限公司 | High-altitude operation safety belt hanging rope high-hanging low-hanging use identification method and device and electronic equipment |
CN114694073A (en) * | 2022-04-06 | 2022-07-01 | 广东律诚工程咨询有限公司 | Intelligent detection method and device for wearing condition of safety belt, storage medium and equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN209060402U (en) * | 2018-10-16 | 2019-07-05 | 三峡大学 | A kind of detection is high to hang low safety belt |
CN110404202A (en) * | 2019-06-28 | 2019-11-05 | 北京市政建设集团有限责任公司 | The detection method and device of aerial work safety belt, aerial work safety belt |
CN110517448A (en) * | 2019-09-02 | 2019-11-29 | 成都建工第五建筑工程有限公司 | High-altitude safe operational method and seat belt hanger detection system for high altitude operation |
CN110533076A (en) * | 2019-08-01 | 2019-12-03 | 江苏濠汉信息技术有限公司 | The detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis |
CN111144263A (en) * | 2019-12-20 | 2020-05-12 | 山东大学 | Construction worker high-fall accident early warning method and device |
-
2020
- 2020-10-12 CN CN202011082084.XA patent/CN112215138A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN209060402U (en) * | 2018-10-16 | 2019-07-05 | 三峡大学 | A kind of detection is high to hang low safety belt |
CN110404202A (en) * | 2019-06-28 | 2019-11-05 | 北京市政建设集团有限责任公司 | The detection method and device of aerial work safety belt, aerial work safety belt |
CN110533076A (en) * | 2019-08-01 | 2019-12-03 | 江苏濠汉信息技术有限公司 | The detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis |
CN110517448A (en) * | 2019-09-02 | 2019-11-29 | 成都建工第五建筑工程有限公司 | High-altitude safe operational method and seat belt hanger detection system for high altitude operation |
CN111144263A (en) * | 2019-12-20 | 2020-05-12 | 山东大学 | Construction worker high-fall accident early warning method and device |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113392708A (en) * | 2021-05-13 | 2021-09-14 | 上海湃道智能科技有限公司 | Safety belt detection method |
CN114155492A (en) * | 2021-12-09 | 2022-03-08 | 华电宁夏灵武发电有限公司 | High-altitude operation safety belt hanging rope high-hanging low-hanging use identification method and device and electronic equipment |
CN114694073A (en) * | 2022-04-06 | 2022-07-01 | 广东律诚工程咨询有限公司 | Intelligent detection method and device for wearing condition of safety belt, storage medium and equipment |
CN114694073B (en) * | 2022-04-06 | 2023-06-06 | 广东律诚工程咨询有限公司 | Intelligent detection method, device, storage medium and equipment for wearing condition of safety belt |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112215138A (en) | Deep learning-based violation detection method for low hanging height of safety belt | |
CN110070033B (en) | Method for detecting wearing state of safety helmet in dangerous working area in power field | |
CN111144263A (en) | Construction worker high-fall accident early warning method and device | |
CN110414400B (en) | Automatic detection method and system for wearing of safety helmet on construction site | |
CN109002801B (en) | Face shielding detection method and system based on video monitoring | |
CN111242004A (en) | Automatic alarm method and system based on elevator monitoring data processing | |
CN109389040B (en) | Inspection method and device for safety dressing of personnel in operation field | |
CN106263213A (en) | Safety helmet based on facial recognition techniques wears supervisory systems and method | |
CN111601081A (en) | Method and device for monitoring operation of hanging basket | |
CN113989711B (en) | Power distribution construction safety tool use identification method and system | |
CN112200108A (en) | Mask face recognition method | |
CN113506416A (en) | Engineering abnormity early warning method and system based on intelligent visual analysis | |
CN116259002A (en) | Human body dangerous behavior analysis method based on video | |
CN114155492A (en) | High-altitude operation safety belt hanging rope high-hanging low-hanging use identification method and device and electronic equipment | |
CN114092875A (en) | Operation site safety supervision method and device based on machine learning | |
CN112101180B (en) | Method and system for identifying unsafe behaviors of people | |
CN113688759A (en) | Safety helmet identification method based on deep learning | |
CN113536842A (en) | Electric power operator safety dressing identification method and device | |
CN115690687A (en) | Safe wearing standard detection system based on deep learning technology | |
CN115620340A (en) | Intelligent unsafe behavior identification method based on YOLO | |
CN114359831A (en) | Risk omen reasoning-oriented intelligent identification system and method for worker side-falling | |
CN111274888B (en) | Helmet and work clothes intelligent identification method based on wearable mobile glasses | |
Zhang et al. | The method for recognizing recognition helmet based on color and shape | |
CN111126405B (en) | Real-time scanning and updating pushing method for monitoring fault alarm information of converter station | |
CN113610759A (en) | A on-spot safe management and control system for roofbolter construction |
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 | ||
CB03 | Change of inventor or designer information | ||
CB03 | Change of inventor or designer information |
Inventor after: Zhang Weishan Inventor after: Feng Zhizhen Inventor after: Yu Qiang Inventor before: Feng Zhizhen Inventor before: Zhang Weishan Inventor before: Yu Qiang |
|
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210112 |