CN110533076A - The detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis - Google Patents
The detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis Download PDFInfo
- Publication number
- CN110533076A CN110533076A CN201910708229.3A CN201910708229A CN110533076A CN 110533076 A CN110533076 A CN 110533076A CN 201910708229 A CN201910708229 A CN 201910708229A CN 110533076 A CN110533076 A CN 110533076A
- Authority
- CN
- China
- Prior art keywords
- construction personnel
- seatbelt wearing
- human body
- torso area
- safety belt
- 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
- 238000010276 construction Methods 0.000 title claims abstract description 84
- 238000001514 detection method Methods 0.000 title claims abstract description 35
- 238000004458 analytical method Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 claims abstract description 12
- 238000013145 classification model Methods 0.000 claims abstract description 9
- 238000012544 monitoring process Methods 0.000 claims abstract description 8
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 15
- 238000013526 transfer learning Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 abstract description 7
- 230000000007 visual effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis, method includes: the image pattern for acquiring construction personnel, constitute sample set, the torso area image for intercepting construction personnel, is divided into wear safety belt and not wear safety belt for torso area image;On sample set, using VGG train classification models, seatbelt wearing disaggregated model is obtained;The live video stream of construction site monitoring camera is acquired, extraction obtains key frame images;The key frame images are analyzed using OpenPose human body attitude algorithm for estimating, judge whether to detect human body key point, if so, the torso area of positioning human body;Call seatbelt wearing disaggregated model to torso area row Classification and Identification, judge construction personnel whether wear safety belt.The present invention is able to achieve round-the-clock construction personnel's abnormal activity detection in 24 hours and improves construction efficiency and safety to realize the security control of construction site.
Description
Technical field
The present invention relates to safe construction field, in particular to a kind of construction personnel's seatbelt wearing of view-based access control model analysis
Detection method and device.
Background technique
Safe construction is the cardinal task in power construction, and it is extremely important to carry out site operation safety measure.Construction personnel
Abnormal activity be generate construction safety risk one of principal element, such as: the non-wear safety belt of construction personnel is not advised
Model behavior etc. is all the factor for generating construction safety risk.The safety management of existing power construction mainly or by manpower is supervised
It superintends and directs, be easy to cause supervision loophole, influence construction efficiency, and there are safety issues.
Summary of the invention
The technical problem to be solved in the present invention is that in view of the above drawbacks of the prior art, providing one kind, to be able to achieve 24 small
When round-the-clock construction personnel's abnormal activity detection to realize the security control of construction site improve construction efficiency and peace
The detection method and device of construction personnel's seatbelt wearing of the view-based access control model analysis of full property.
The technical solution adopted by the present invention to solve the technical problems is: constructing a kind of construction personnel of view-based access control model analysis
The detection method of seatbelt wearing, includes the following steps:
A the image pattern for) acquiring construction personnel, constitutes sample set, and the body of construction personnel is intercepted from described image sample
The torso area image is divided into wear safety belt and not wear safety belt by dry area image;
B) on the sample set, using VGG train classification models, seatbelt wearing disaggregated model is obtained;
C the live video stream of construction site monitoring camera) is acquired, extraction obtains key frame images;
D the key frame images) are analyzed using OpenPose human body attitude algorithm for estimating, judge whether to detect that human body closes
Key point, if so, executing step E);Otherwise, step G is executed);
E the torso area of human body) is positioned;
F) the seatbelt wearing disaggregated model is called to carry out Classification and Identification to the torso area, judges the constructor
Member whether wear safety belt;
G) terminate this detection.
In the detection method of construction personnel's seatbelt wearing of view-based access control model of the present invention analysis, the step B)
Further comprise:
B1) image pattern in the sample set is pre-processed, the pretreatment includes at least Random-Rotation, cross
It moves, scale and overturns;
B2) the VGG Net of building setting layer, and download the pre-training model on MSCOCO data set;
B3 described image sample) is loaded on the pre-training model using transfer learning method, every setting step output
Accuracy rate, until being optimal as a result, saving the seatbelt wearing disaggregated model.
In the detection method of construction personnel's seatbelt wearing of view-based access control model of the present invention analysis, the setting layer
It is 19 layers.
In the detection method of construction personnel's seatbelt wearing of view-based access control model of the present invention analysis, the setting step
For 50 steps.
The invention further relates to a kind of detection methods of construction personnel's seatbelt wearing for realizing above-mentioned view-based access control model analysis
Device, comprising:
Image pattern acquisition unit: for acquiring the image pattern of construction personnel, sample set is constituted, from described image sample
The torso area image of middle interception construction personnel, is divided into wear safety belt and not safe wearing for the torso area image
Band;
Seatbelt wearing disaggregated model acquiring unit: for being obtained in the sample set using VGG train classification models
To seatbelt wearing disaggregated model;
Key frame images acquiring unit: for acquiring the live video stream of construction site monitoring camera, extraction is closed
Key frame image;
Human body key point judging unit: for analyzing the key frame figure using OpenPose human body attitude algorithm for estimating
Picture judges whether to detect human body key point;
Torso area positioning unit: for positioning the torso area of human body;
Wear safety belt judging unit: for calling the seatbelt wearing disaggregated model to divide the torso area
Class identification, judge the construction personnel whether wear safety belt;
End unit: for terminating this detection.
In device of the present invention, the seatbelt wearing disaggregated model acquiring unit further comprises:
Image pattern preprocessing module: for the image pattern in the sample set to be pre-processed, the pretreatment
Including at least Random-Rotation, traversing, scaling and overturning;
Model download module: for constructing the VGG Net of setting layer, and the pre-training mould on MSCOCO data set is downloaded
Type;
Image pattern loading module: for loading described image sample on the pre-training model using transfer learning method
This, walks output accuracy rate every setting, until being optimal as a result, saving the seatbelt wearing disaggregated model.
In device of the present invention, the layer that sets is 19 layer.
In device of the present invention, the setting step is 50 steps.
The detection method and device for implementing construction personnel's seatbelt wearing of view-based access control model analysis of the invention, have following
The utility model has the advantages that being instructed since the torso area image of construction personnel is divided into wear safety belt and not wear safety belt using VGG
Practice disaggregated model, obtains seatbelt wearing disaggregated model;It is extracted from the live video stream of acquisition and obtains key frame images, utilized
OpenPose human body attitude algorithm for estimating analysis of key frame image positions human body key point, positions the torso area of human body;Safety
Band wear disaggregated model to torso area carry out Classification and Identification, judge construction personnel whether wear safety belt;The present invention is able to achieve
Round-the-clock construction personnel's abnormal activity detection in 24 hours improves construction efficiency to realize the security control of construction site
And safety.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is that the present invention is based on detection method and device one embodiment of construction personnel's seatbelt wearing of visual analysis
The flow chart of middle method;
Fig. 2 is, using VGG train classification models, to obtain seatbelt wearing classification mould in the embodiment on sample set
The specific flow chart of type;
Fig. 3 is the structural schematic diagram of device in the embodiment.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The present invention is based in the detection method and device embodiment of construction personnel's seatbelt wearing of visual analysis, base
It is as shown in Figure 1 in the flow chart of the detection method of construction personnel's seatbelt wearing of visual analysis.In Fig. 1, the view-based access control model point
The detection method of construction personnel's seatbelt wearing of analysis includes the following steps:
Step S01 acquires the image pattern of construction personnel, constitutes sample set, and the body of construction personnel is intercepted from image pattern
Torso area image is divided into wear safety belt and not wear safety belt by dry area image: in this step, acquiring construction personnel
Image pattern, these image patterns constitute sample set, from image pattern intercept construction personnel torso area image (also
It is topography of the screenshot in relation to torso area from the image pattern of construction personnel), which is divided into two
Class, one type are wear safety belts, and another kind of is not wear safety belt.
Step S02, using VGG train classification models, obtains seatbelt wearing disaggregated model: this step on sample set
In, on step S01 sample set collected, using VGG train classification models, obtain seatbelt wearing disaggregated model Mh.
Step S03 acquires the live video stream of construction site monitoring camera, and extraction obtains key frame images: this step
In, the live video stream of the monitoring camera of acquisition arrangement at the construction field (site) therefrom extracts key frame images I.
Step S04 utilizes OpenPose human body attitude algorithm for estimating analysis of key frame image, judges whether to detect human body
Key point: in this step, using obtained key frame images I in OpenPose human body attitude algorithm for estimating analytical procedure S03,
Position human body key point, judge whether to detect human body key point, if it is determined that result be it is yes, then follow the steps S05;It is no
Then, step S07 is executed.
The torso area of step S05 positioning human body: in this step, positioning human body key point, obtain the torso area of human body,
Also just the torso area of human body is positioned.
Step S06 calls seatbelt wearing disaggregated model to carry out Classification and Identification to torso area, whether judges construction personnel
Wear safety belt: in this step, the seatbelt wearing disaggregated model Mh that training obtains in invocation step S02 is obtained in step S05
The torso area that arrives carries out Classification and Identification, judge the construction personnel whether wear safety belt.
Step S07 terminates this detection: in this step, terminating this detection.
The detection method of construction personnel's seatbelt wearing of view-based access control model analysis of the invention has liberated the double of supervisor
Eye has been able to achieve round-the-clock construction personnel's abnormal activity detection in 24 hours, to realize the security control of construction site, has mentioned
High construction efficiency and safety.
For the present embodiment, above-mentioned steps S02 can also be refined further, and the flow chart after refinement is as shown in Figure 2.
In Fig. 2, above-mentioned steps S02 further comprises following steps:
Step S21 pre-processes the image pattern in sample set, and pretreatment includes at least Random-Rotation, traversing, contracting
Put and overturn: in this step, the image pattern in sample set being pre-processed, pretreatment include at least Random-Rotation, it is traversing,
Scaling and overturning etc. carry out data enhancing using the methods of Random-Rotation, traversing, scaling, overturning in this way, expand image pattern amount
The VGG Net of step S22 building setting layer, and download the pre-training model on MSCOCO data set: in this step,
The VGG Net of building setting layer, and download the pre-training model on MSCOCO data set.In the present embodiment, this sets layer as 19
Layer, certainly, in practical applications, the number of plies of the setting layer can also be increase accordingly or be reduced according to specific requirements.
Step S23 uses transfer learning method load image sample on pre-training model, accurate every setting step output
Rate, until being optimal as a result, saving seatbelt wearing disaggregated model: in this step, using transfer learning method in pre-training
Load image sample on model walks output accuracy rate every setting, until being optimal as a result, saving seatbelt wearing classification mould
Type Mh.In the present embodiment, setting step is 50 steps, and certainly, in practical applications, the size of setting step can be according to specific need
It asks and is increase accordingly or reduced.
The present embodiment further relates to a kind of detection method of construction personnel's seatbelt wearing for realizing above-mentioned view-based access control model analysis
Device, the apparatus structure schematic diagram is as shown in Figure 3.In Fig. 3, which includes image pattern acquisition unit 1, seatbelt wearing
Disaggregated model acquiring unit 2, key frame images acquiring unit 3, human body key point judging unit 4, torso area positioning unit 5,
Wear safety belt judging unit 6 and end unit 7.Wherein, image pattern acquisition unit 1 is used to acquire the image sample of construction personnel
This, these image patterns constitute sample set, the torso area image of construction personnel are intercepted from image pattern, by torso area figure
As being divided into wear safety belt and not wear safety belt.Seatbelt wearing disaggregated model acquiring unit 2 is used in sample set, benefit
With VGG train classification models, seatbelt wearing disaggregated model Mh is obtained;Key frame images acquiring unit 3 is existing for acquiring construction
The live video stream of field monitoring camera, extraction obtain key frame images I;Human body key point judging unit 4 is for utilizing
OpenPose human body attitude algorithm for estimating analysis of key frame image I, judges whether to detect human body key point;Torso area positioning
Unit 5 is used to position the torso area of human body;Wear safety belt judging unit 6 is for calling Mh pairs of seatbelt wearing disaggregated model
Shi worker person Alto torso area carry out Classification and Identification, judge construction personnel whether wear safety belt;End unit 7 is for terminating this
Secondary detection.
The device of the invention has liberated the eyes of supervisor, and it is lack of standardization to be able to achieve 24 hours round-the-clock construction personnel
Behavioral value improves construction efficiency and safety to realize the security control of construction site.
In the present embodiment, seatbelt wearing disaggregated model acquiring unit 2 further comprise image pattern preprocessing module 21,
Model download module 22 and image pattern loading module 23;Wherein, image pattern preprocessing module 21 is for will be in sample set
Image pattern is pre-processed, and pretreatment includes at least Random-Rotation, traversing, scaling and overturning.
Model download module 22 is used to construct the VGG Net of setting layer, and downloads the pre-training mould on MSCOCO data set
Type.In the present embodiment, this sets layer as 19 layers, and certainly, in practical applications, the number of plies of the setting layer can also be according to specific need
It asks and is increase accordingly or reduced.
Image pattern loading module 23 is used to use transfer learning method load image sample on pre-training model, every
Setting step output accuracy rate, until being optimal as a result, saving seatbelt wearing disaggregated model.In the present embodiment, setting step
For 50 steps, certainly, in practical applications, the size of setting step can be increase accordingly or be reduced according to specific requirements.
In short, the present invention is directed to the abnormal activity of the non-wear safety belt of construction personnel, propose based on intelligent vision point
The detection method of analysis technology has liberated the eyes of supervisor, is able to achieve round-the-clock construction personnel's abnormal activity in 24 hours
Detection improves construction efficiency and safety to realize the security control of construction site.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of detection method of construction personnel's seatbelt wearing of view-based access control model analysis, which comprises the steps of:
A the image pattern for) acquiring construction personnel, constitutes sample set, and the trunk area of construction personnel is intercepted from described image sample
The torso area image is divided into wear safety belt and not wear safety belt by area image;
B) on the sample set, using VGG train classification models, seatbelt wearing disaggregated model is obtained;
C the live video stream of construction site monitoring camera) is acquired, extraction obtains key frame images;
D the key frame images) are analyzed using OpenPose human body attitude algorithm for estimating, judge whether to detect human body key
Point, if so, executing step E);Otherwise, step G is executed);
E the torso area of human body) is positioned;
F) the seatbelt wearing disaggregated model is called to carry out Classification and Identification to the torso area, judges that the construction personnel is
No wear safety belt;
G) terminate this detection.
2. the detection method of construction personnel's seatbelt wearing of view-based access control model analysis according to claim 1, feature exist
Further comprise in the step B):
B1) image pattern in the sample set is pre-processed, the pretreatment includes at least Random-Rotation, traversing, contracting
It puts and overturns;
B2) the VGG Net of building setting layer, and download the pre-training model on MSCOCO data set;
B3 described image sample) is loaded on the pre-training model using transfer learning method, it is accurate every setting step output
Rate, until being optimal as a result, saving the seatbelt wearing disaggregated model.
3. the detection method of construction personnel's seatbelt wearing of view-based access control model analysis according to claim 2, feature exist
In the layer that sets is 19 layer.
4. the detection method of construction personnel's seatbelt wearing of view-based access control model analysis according to claim 2, feature exist
In the setting step is 50 steps.
5. a kind of dress of the detection method for the construction personnel's seatbelt wearing for realizing view-based access control model analysis as described in claim 1
It sets characterized by comprising
Image pattern acquisition unit: for acquiring the image pattern of construction personnel, sample set is constituted, is cut from described image sample
The torso area image is divided into wear safety belt and not wear safety belt by the torso area image for taking construction personnel;
Seatbelt wearing disaggregated model acquiring unit: for being pacified in the sample set using VGG train classification models
Full band wears disaggregated model;
Key frame images acquiring unit: for acquiring the live video stream of construction site monitoring camera, extraction obtains key frame
Image;
Human body key point judging unit: for analyzing the key frame images using OpenPose human body attitude algorithm for estimating, sentence
It is disconnected whether to detect human body key point;
Torso area positioning unit: for positioning the torso area of human body;
Wear safety belt judging unit: for calling the seatbelt wearing disaggregated model to carry out classification knowledge to the torso area
, do not judge the construction personnel whether wear safety belt;
End unit: for terminating this detection.
6. device according to claim 5, which is characterized in that the seatbelt wearing disaggregated model acquiring unit is further
Include:
Image pattern preprocessing module: for pre-processing the image pattern in the sample set, the pretreatment is at least
Including Random-Rotation, traversing, scaling and overturning;
Model download module: for constructing the VGG Net of setting layer, and the pre-training model on MSCOCO data set is downloaded;
Image pattern loading module: for loading described image sample on the pre-training model using transfer learning method,
Output accuracy rate is walked every setting, until being optimal as a result, saving the seatbelt wearing disaggregated model.
7. device according to claim 6, which is characterized in that the layer that sets is 19 layer.
8. device according to claim 6, which is characterized in that the setting step is 50 steps.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910708229.3A CN110533076A (en) | 2019-08-01 | 2019-08-01 | The detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910708229.3A CN110533076A (en) | 2019-08-01 | 2019-08-01 | The detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110533076A true CN110533076A (en) | 2019-12-03 |
Family
ID=68661980
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910708229.3A Pending CN110533076A (en) | 2019-08-01 | 2019-08-01 | The detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110533076A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111191705A (en) * | 2019-12-24 | 2020-05-22 | 上海眼控科技股份有限公司 | Method, apparatus and computer-readable storage medium for detecting safety equipment of human body |
CN112215138A (en) * | 2020-10-12 | 2021-01-12 | 中国石油大学(华东) | Deep learning-based violation detection method for low hanging height of safety belt |
CN112257580A (en) * | 2020-10-21 | 2021-01-22 | 中国石油大学(华东) | Human body key point positioning detection method based on deep learning |
CN112766070A (en) * | 2020-12-31 | 2021-05-07 | 上海电机学院 | Intelligent detection method and system for wearing of labor protection protective articles of electrical overhaul personnel |
CN113255509A (en) * | 2021-05-20 | 2021-08-13 | 福州大学 | Building site dangerous behavior monitoring method based on Yolov3 and OpenPose |
CN113823180A (en) * | 2021-09-29 | 2021-12-21 | 国网山东省电力公司临朐县供电公司 | Signboard hanging device and method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108038424A (en) * | 2017-11-27 | 2018-05-15 | 华中科技大学 | A kind of vision automated detection method suitable for working at height |
CN108596138A (en) * | 2018-05-03 | 2018-09-28 | 南京大学 | A kind of face identification method based on migration hierarchical network |
CN109255312A (en) * | 2018-08-30 | 2019-01-22 | 罗普特(厦门)科技集团有限公司 | A kind of abnormal dressing detection method and device based on appearance features |
-
2019
- 2019-08-01 CN CN201910708229.3A patent/CN110533076A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108038424A (en) * | 2017-11-27 | 2018-05-15 | 华中科技大学 | A kind of vision automated detection method suitable for working at height |
CN108596138A (en) * | 2018-05-03 | 2018-09-28 | 南京大学 | A kind of face identification method based on migration hierarchical network |
CN109255312A (en) * | 2018-08-30 | 2019-01-22 | 罗普特(厦门)科技集团有限公司 | A kind of abnormal dressing detection method and device based on appearance features |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111191705A (en) * | 2019-12-24 | 2020-05-22 | 上海眼控科技股份有限公司 | Method, apparatus and computer-readable storage medium for detecting safety equipment of human body |
CN112215138A (en) * | 2020-10-12 | 2021-01-12 | 中国石油大学(华东) | Deep learning-based violation detection method for low hanging height of safety belt |
CN112257580A (en) * | 2020-10-21 | 2021-01-22 | 中国石油大学(华东) | Human body key point positioning detection method based on deep learning |
CN112766070A (en) * | 2020-12-31 | 2021-05-07 | 上海电机学院 | Intelligent detection method and system for wearing of labor protection protective articles of electrical overhaul personnel |
CN113255509A (en) * | 2021-05-20 | 2021-08-13 | 福州大学 | Building site dangerous behavior monitoring method based on Yolov3 and OpenPose |
CN113823180A (en) * | 2021-09-29 | 2021-12-21 | 国网山东省电力公司临朐县供电公司 | Signboard hanging device and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110533076A (en) | The detection method and device of construction personnel's seatbelt wearing of view-based access control model analysis | |
CN111144263B (en) | Construction worker high-falling accident early warning method and device | |
CN105310696B (en) | A kind of fall detection model building method and corresponding fall detection method and device | |
CN109034215A (en) | A kind of safety cap wearing detection method based on depth convolutional neural networks | |
CN109711297A (en) | Risk Identification Method, device, computer equipment and storage medium based on facial picture | |
CN106919921B (en) | Gait recognition method and system combining subspace learning and tensor neural network | |
CN108710126A (en) | Automation detection expulsion goal approach and its system | |
CN110321809A (en) | A kind of substation's operation field monitoring method and device based on deep learning | |
CN108683724A (en) | A kind of intelligence children's safety and gait health monitoring system | |
CN110569722A (en) | Visual analysis-based constructor dressing standard detection method and device | |
CN108062542A (en) | The detection method for the face being blocked | |
CN109684981A (en) | Glaucoma image-recognizing method, equipment and screening system | |
CN109241842A (en) | Method for detecting fatigue driving, device, computer equipment and storage medium | |
KR102411316B1 (en) | Method for providing worker safty analysis service in construction site and server and system for performing the same | |
CN109034134A (en) | Abnormal driving behavioral value method based on multitask depth convolutional neural networks | |
CN110096983A (en) | The safe dress ornament detection method of construction worker in a kind of image neural network based | |
CN108958482B (en) | Similarity action recognition device and method based on convolutional neural network | |
CN103942539A (en) | Method for accurately and efficiently extracting human head ellipse and detecting shielded human face | |
CN109697716A (en) | Glaucoma image-recognizing method, equipment and screening system | |
CN111259855A (en) | Mobile safety helmet wearing detection method based on deep learning | |
CN102457705A (en) | Method and system for detecting and monitoring fight behavior | |
CN110532838A (en) | Object test equipment and method and storage medium | |
CN113050478A (en) | Method for monitoring and early warning operation state of transformer substation constructor | |
JP2019109760A (en) | Image monitoring device | |
CN116580490A (en) | Wind power operation safety detection method and system |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191203 |