CN113506416A - Engineering abnormity early warning method and system based on intelligent visual analysis - Google Patents
Engineering abnormity early warning method and system based on intelligent visual analysis Download PDFInfo
- Publication number
- CN113506416A CN113506416A CN202110753861.7A CN202110753861A CN113506416A CN 113506416 A CN113506416 A CN 113506416A CN 202110753861 A CN202110753861 A CN 202110753861A CN 113506416 A CN113506416 A CN 113506416A
- Authority
- CN
- China
- Prior art keywords
- visual analysis
- worker
- intelligent visual
- platform
- early warning
- 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
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/08—Alarms for ensuring the safety of persons responsive to the presence of persons in a body of water, e.g. a swimming pool; responsive to an abnormal condition of a body of water
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Alarm Systems (AREA)
Abstract
The invention relates to the technical field of water conservancy security engineering, in particular to an engineering abnormity early warning method and system based on intelligent visual analysis, wherein the system comprises a client platform, a data transmission platform and an intelligent visual analysis platform; the client accesses the intelligent visual analysis platform locally and remotely by adding a wireless local area network or a broadband mode; the data transmission platform realizes the data transmission between the client and the intelligent visual analysis platform; the intelligent visual analysis platform comprises a monitoring camera, a server and a display, wherein the camera is used for collecting a hydraulic engineering field monitoring video as the input of the system, the server is used for processing images in video data, the operations comprise target recognition, human body posture detection, abnormal behavior judgment and the like, and the processing result is displayed on the display in real time. The invention realizes effective detection and early warning of safety accidents possibly occurring by workers in the hydraulic engineering field, and achieves the effect of preventing the safety accidents.
Description
Technical Field
The invention relates to the technical field of water conservancy security engineering, in particular to an engineering abnormity early warning method and system based on intelligent visual analysis.
Background
Currently, monitoring devices have been widely used, and are found in the fields of transportation, medical treatment, construction sites, and even home residences. Traditional video monitoring is mainly recorded for people to observe, but the method is time-consuming and labor-consuming, has limited video retention time, and is difficult to analyze and observe a large amount of collected data in a limited time. The intelligent video monitoring system has the capability of real-time processing and intelligent analysis, and solves the problem of the defects of manual observation.
In an actual hydraulic engineering site, a lot of sudden problems exist, which not only require a management unit to have a perfect emergency and treatment scheme, but also require personnel entering the hydraulic engineering site to have certain protection measures, such as correctly wearing a life jacket, stably walking and the like. The problems that hydraulic engineering field personnel lack safety worry awareness, engineering field inspection efficiency is low, a complete management system is lacked and the like exist, so that a plurality of safety accidents occur.
In recent years, the country continuously attaches importance to safety production, hydraulic engineering sites have higher and higher requirements on operation specifications of personnel, most hydraulic engineering sites are provided with monitors, and all-weather shooting of the environment and operating personnel of the hydraulic engineering sites can be achieved. For example, Chinese patent CN103456136A discloses a monitoring and early warning system and method for major accident potential safety hazards of Internet of things framework hydraulic and hydro-power engineering, which identify and position potential safety hazard information which may cause potential dangerous accidents in a hydraulic and hydro-power engineering area, determine the type and the grade of the potential safety hazards, perform type judgment, grade judgment and information statistics on the potential safety hazards, analyze the geographical position, the spatial position, the dynamic change, the state prediction and the historical information in the area in combination with the monitoring points in the area, automatically generate early warning information after finding that the potential safety hazard information in the monitoring area exceeds a set warning value, and provide comprehensive data information of the potential safety hazards in the area and provide early warning of the major accident potential safety hazards of the engineering through communication transmission. But the potential safety hazard still exists at the hydraulic engineering scene. With the rapid development of deep learning and the promotion of information management by engineering units, a set of engineering abnormity early warning method and system based on intelligent visual analysis and suitable for hydraulic engineering scenes is urgently needed to be designed.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, provides an engineering abnormity early warning method and system based on intelligent visual analysis, solves the problem that the existing engineering abnormity early warning system is difficult to effectively early warn dangerous accidents which may occur, and can avoid the safety accidents which may occur in engineering.
The technical scheme adopted by the invention for solving the technical problems is as follows:
an engineering abnormity early warning method based on intelligent visual analysis is characterized in that whether a worker correctly wears a life jacket and whether abnormal dangerous actions occur is detected in real time by designing an intelligent analysis module, and the method specifically comprises the following steps:
1) utilizing a monitoring camera to acquire a hydraulic engineering field monitoring video image in real time;
2) detecting life jackets and workers appearing in the images by using a YOLOv4 algorithm;
3) estimating the human body posture of a worker by adopting a human body posture identification algorithm based on HigherHRNet to obtain human body key points;
4) judging whether the worker wears the life jacket or not by combining the coordinate position of the life jacket detected by the YOLOv4 algorithm and the coordinates of the worker;
5) judging whether the worker is correctly fallen and climbed by combining the coordinate position of the worker detected by the YOLOv4 algorithm and the coordinates of the left shoulder, the right shoulder, the waist and the legs of the worker;
6) a dangerous area is marked in the hydraulic engineering area by utilizing a hydraulic engineering field monitoring video image acquired by a monitoring camera in real time, and whether a worker invades the dangerous area is judged by detecting whether the worker appears in the dangerous area.
The invention not only makes the field safety management work simpler and more efficient, but also can give an early warning to remind workers in time, thereby avoiding the occurrence of danger and having great significance for protecting the safety of workers and property.
The invention also provides another technical scheme that,
a system for implementing the intelligent visual analysis-based engineering abnormity early warning method comprises the following steps:
a client platform: connecting a wireless local area network or a broadband, and locally and remotely accessing an intelligent visual analysis platform to realize remote target detection;
a data transmission platform: the system is used for realizing data transmission between the intelligent visual analysis platform and the client platform;
the intelligent visual analysis platform comprises: the system comprises a camera, a server and a display screen, and real-time detection and judgment are carried out on whether a worker correctly wears the life jacket or not by adopting a YOLOv4 algorithm and a human posture recognition algorithm based on HigherHRNet on the acquired hydraulic engineering field monitoring video images.
The invention has the technical effects that:
compared with the prior art, the engineering abnormity early warning system based on intelligent visual analysis has the following advantages:
(1) the invention collects the site image of the hydraulic engineering based on the monitoring camera, can provide the site image in real time, and is convenient for rapidly detecting the construction site;
(2) the invention provides that the target detection is applied to the hydraulic engineering field, the YOLOv4 algorithm has very high real-time performance and accuracy, the human body posture estimation detection rate based on HigherHRNet is high, and the performance is superior under the condition of dense crowd;
(3) the invention adopts the Shiro authority management, only the legal user can use the system, thereby ensuring the safety and the confidentiality of the monitoring data, and simultaneously giving a certain authority to the user who is successfully registered, for example, only the manager can use the management authority. The system is simple to operate, has concise pages, and is convenient for the use of the supervisor.
Drawings
FIG. 1 is a schematic diagram of a platform model of an engineering anomaly early warning system based on intelligent visual analysis according to the invention;
FIG. 2 is a schematic diagram of the software architecture of the intelligent visual analysis platform of the present invention;
FIG. 3 is a schematic diagram illustrating the fall determination in the intelligent visual analysis platform module according to the present invention;
FIG. 4 is a schematic diagram illustrating climbing discrimination in an intelligent visual analysis platform module according to the present invention;
FIG. 5 is a schematic diagram of key points of human body posture in the intelligent visual analysis platform module according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described below with reference to the drawings of the specification.
Example (b):
as shown in fig. 1, the engineering anomaly early warning system based on intelligent visual analysis according to the embodiment includes a client platform, a data transmission platform, and an intelligent visual analysis platform.
The client platform is connected with a wireless local area network or a broadband, comprises a manager, a water conservancy construction site client, a mobile client and the like, and locally and remotely accesses the intelligent visual analysis platform to realize remote target detection.
The data transmission platform realizes the transmission of image data by utilizing a CDMA or GPRS network; the client can locally and remotely access the intelligent visual analysis platform by adding a wireless local area network, so that the remote detection of the hydraulic engineering field is realized.
The intelligent visual analysis platform comprises a monitoring camera, a server and a display screen, and the camera is used for acquiring a hydraulic engineering field monitoring video as the input of the system; in the server, the image in the video data is subjected to operations such as target recognition, human body posture detection and the like according to the instruction, and a result after a series of processing is displayed on a display screen in real time.
As shown in fig. 2, a schematic diagram of a software structure of an intelligent visual analysis platform is shown, the present invention designs server detection software for detecting abnormal behaviors occurring in a hydraulic engineering, which is used for controlling monitoring video information of an engineering monitoring camera and an abnormal behavior detection result, and the result is displayed in a software interface and stored in a database. On the basis of fully analyzing the functions and structures required by the system, the main function modules of the detection software are divided, and the main function modules mainly comprise a user management module, a data management module, an abnormality detection module and an alarm module.
The user management module mainly comprises user registration and user login, wherein a user using the system for the first time needs to register, and can log in the system after the administrator confirms and gives authority.
And the data management module is mainly used for importing the monitoring video data of the construction site, managing and storing the monitoring video data and transmitting the video to the following process. The backup storage of the monitoring video frequency division time point can be realized in the process, so that the later calling is facilitated.
The utility model discloses a life vest, including the life vest, the detection module is used for monitoring the life vest, including the key point that the life vest was worn to the workman, the detection module is used for the key point that the human gesture was estimated and was discerned carries out unusual behavior detection, including falling down, climbing and danger area intrusion behavior, the real-time supervision water conservancy building scene will be shown in the display screen, report to the police and carry out the screenshot and preserve in the database folder through alarm module when appearing unusually simultaneously, it is concrete, whether the workman that appears in the real-time supervision video of anomaly detection module is correctly dressed the life vest and whether take place to fall down, the climbing, the invasion of danger area behavior, include and detect workman and life vest, if there are workman and life vest then combine subsequent human gesture to carry out life vest wearing to detect, and carry out unusual behavior detection through the key point that human gesture estimation discerned, including falling down, climbing and danger area invasion behavior.
The abnormity detection module mainly comprises a life jacket wearing detection module and an abnormal dangerous action detection module; the life jacket wearing detection module acquires a hydraulic engineering field monitoring video image in real time, performs life jacket target identification on the image by adopting a YOLOv4 algorithm, and judges whether a worker wears the life jacket correctly by combining a human posture identification algorithm based on HigherHRNet; the abnormal dangerous action detection module comprises a falling detection module, a climbing detection module and a dangerous area detection module.
The falling detection module and the climbing detection module judge whether the worker falls or not and whether climbing behavior occurs or not according to the relative position relation between the target coordinate position detected in the image and the posture key point; the dangerous area detection module is used for detecting workers through a dangerous area marked in a hydraulic engineering area, so that whether the workers invade the dangerous area or not is judged.
The position information and the category identification information in the image are output as a detection model through an object detection algorithm, and the category and the coordinates of the object are recorded in the output information. The human body posture estimation based on the HigherHRNet takes human body posture skeleton information as output, and the position information records the coordinates of each key point and the posture of the human body.
a) Life jacket wearing determination:
firstly, an object detection algorithm data set is manufactured, wherein the data set mainly comprises two categories of workers and life jackets, when the Yolov4 is detected, frames can be classified according to the workers and the life jackets, the center of a single human posture key point is calculated according to the skeleton point coordinates of a posture detection algorithm, then the human body frame predicted by the Yolov4 is matched with the human body posture detected by the human body posture estimation based on HigherHRNet, and the maximum intersection area ratio IOA of each detected rectangular frame of the life jacket and the corresponding frame of the workers is calculated:
in the formula, SjacketIs the area of the lifejacket frame, SpersonIs the area of the border of the figure, Sjacket∩SpersonThe formula represents the proportion of the intersection of the life jacket (jack) frame and the character frame (person) in the life jacket frame. If this lifejacket rim is present and the IOA is equal to or greater than 0.5, there is a possibility that the worker wears the lifejacket, otherwise it is not.
b) Tumble determination:
as shown in fig. 3, a schematic diagram of fall discrimination in the intelligent visual analysis platform module is shown, and whether a worker falls or not is determined by using 18 key point positions output by human posture estimation based on highherhrnet and a certain vector position relationship.
(1) Judgment step 1
In order to detect the moving speed v capable of having good stability, three key points of 0,1 and 8 in the table 1 are selected, the coordinates of the three corresponding key points in the image are recorded, the coordinate of the human body centroid point is the average value of the three coordinates, and the moving speed v of the centroid point vertical coordinate is calculated once every 10 frames. Setting a certain threshold vthIf v is greater than or equal to vthIf yes, the judgment condition in the judgment step 1 is met, and the v expression is as follows:
wherein i belongs to {0,1,8 };
the ordinate value of the ith key point at the 10 th frame;
the ordinate value of the ith key point in the 1 st frame; t generationTime of the watch, t10-t1Representing the time taken between frame 1 and frame 10.
TABLE 1 human gesture Key Point Numbers and Contents labeled in FIG. 5 of Intelligent visual analysis platform Module
(2) Judgment step 2
It is far from sufficient to judge whether the worker falls in the judging step 1), for example, workers crouch down to pick up article materials or tie shoelaces and the like cause misjudgment, so whether the workers fall or not needs to be further judged. When the worker squats, the maximum height value of the key point 1 in the input image is calculated and recorded as HthAnd v is not less than v in the coincidence judgment step 1thThen, y is recorded every 10 frames1If y is satisfied for a certain period of time1≥HthThe worker may fall down.
Note that HthWill vary with the distance of the person from the camera. Because the regional surveillance camera machine of water conservancy building management and control is generally far away from the workman, the Euclidean distance l between key point 1 and 8 is less, and the expression of l is:
x1is the abscissa value, x, of the 1 st key point8The abscissa value of the 8 th key point; let the distance between the worker and the camera be L, L, L and HthThe three satisfy a certain linear relation and can establish a linear regression model. The relationship between L and L is:
L(li)＝αli+β (4)
wherein i represents the ith time. The optimal values of α and β can be solved using a least squares estimation algorithm, as shown in equations 5 to 8 below.
Wherein
m is a preset value and represents that the average Euclidean distance is calculated every m times
The same can be obtained by obtaining L and HthIs determined by the relationship of (1)thThe value of (c). The process of judging step 2 proves that the worker has a squatting type action, but cannot prove that the worker falls down, sits down quickly or crouches down for a long time for rest, so that further evidence is needed for judging that the worker falls down.
(3) Judgment step 3
The most obvious feature of the fall behavior is that the back of the human body is inclined or even parallel to the ground, and the emphasis is placed on the key points of the left and right shoulders (key points 2 and 5) and the left and right waists (key points 8 and 11).
Taking the difference Δ y between the ordinate of the two key points, if a fall occurs, the value of Δ y should be very small. So the condition is required to be met when the fall is judged:
Δy＝|y2-y8|＜h (9)
in the formula, H is the falling height of the normal person, and H is the bending height of the normal person.
To set the value of H, twenty hydraulic engineering surveillance videos were observed, where the distance of worker keypoint 1 to keypoint 8, i.e., H, appearing in the video is approximately 1/2 of the distance of keypoint 1 to keypoint 10, and H is approximately 2/5 of the distance of keypoint 1 to keypoint 10, i.e.:
h＝2(y10-y1)/5 (10)
c) the climbing determination:
as shown in fig. 4, which is a schematic diagram of climbing discrimination in the intelligent visual analysis platform module according to the present invention, whether a worker climbs is determined by using 18 key point positions output by human posture estimation based on highherhrnet according to a certain vector position relationship.
First, it is determined that there is a certain foot lifting action when climbing, and one foot lifts first, and the right foot lifts first in the following, when y is13＜y10When the worker walks or climbs, the worker starts to walk or climb.
(1) Judgment step 1
First consider the ordinate y of the right knee of a worker9Whether or not it is greater than the ordinate y of the right hip8If y is8＜y9And the right leg of the worker is in the climbing state. But do not represent y8≥y9In time, the worker is not climbing, so further determination is needed.
(2) Judgment step 2
When y is8≥y9When the worker climbs, the central point of the human body moves upwards. Note CtIs the center of the target p at time t, CtVector difference from previous time t-k is VtIf V istFor the ascent state, it can be judged that the worker is in the climbing state. To distinguish the worker from a jumping or climbing situation, VtNeed to be greater than a certain threshold TjTherefore, the conditions required for climbing are as follows:
wherein the content of the first and second substances,
is the y-axis component of target p at time t; t isjSet to small jump VtThe rise amount of (2) is scientifically 50 cm in height for a normal person to jump at one time, and is set here as 1/3 the height of a worker in motion for the generality of the experimental results.
d) Intrusion determination in hazardous areas
A plurality of river areas exist in a hydraulic engineering field, and a plurality of dangers exist, so that special cameras are required to be installed on roads on two sides of a dangerous area, a green dangerous area is divided, and if workers are detected in the area, the workers are judged to be dangerous invasion.
And the alarm module automatically detects an alarm and records and archives specific condition when a worker is detected but a life jacket is not detected, the worker and the life jacket are detected but the worker does not wear the life jacket, the worker falls down, the worker climbs and the worker invades a dangerous area.
The above embodiments are only specific examples of the present invention, and the protection scope of the present invention includes but is not limited to the product forms and styles of the above embodiments, and any suitable changes or modifications made by those skilled in the art according to the claims of the present invention shall fall within the protection scope of the present invention.
Claims (5)
1. An engineering abnormity early warning method based on intelligent visual analysis is characterized in that: through designing intelligent analysis module, whether correctly dress the life vest and whether take place unusual dangerous action to the workman and carry out real-time detection, specifically include the following step:
1) utilizing a monitoring camera to acquire a hydraulic engineering field monitoring video image in real time;
2) detecting life jackets and workers appearing in the images by using a YOLOv4 algorithm;
3) estimating the human body posture of a worker by adopting a human body posture identification algorithm based on HigherHRNet to obtain human body key points;
4) judging whether the worker wears the life jacket or not by combining the coordinate position of the life jacket detected by the YOLOv4 algorithm and the coordinates of the worker;
5) judging whether the worker is correctly fallen and climbed by combining the coordinate position of the worker detected by the YOLOv4 algorithm and the coordinates of the left shoulder, the right shoulder, the waist and the legs of the worker;
6) a dangerous area is marked in the hydraulic engineering area by utilizing a hydraulic engineering field monitoring video image acquired by a monitoring camera in real time, and whether a worker invades the dangerous area is judged by detecting whether the worker appears in the dangerous area.
2. The utility model provides an engineering abnormity early warning system based on intelligent visual analysis which characterized in that: method for implementing claim 1, comprising:
the client platform is connected with a wireless local area network or a broadband, and locally and remotely accesses the intelligent visual analysis platform to realize remote target detection;
the data transmission platform is used for realizing data transmission between the intelligent visual analysis platform and the client platform;
the intelligent visual analysis platform comprises a camera, a server and a display screen, and is used for detecting and distinguishing whether workers wear life jackets correctly in real time or not by using a YOLOv4 algorithm and a human posture recognition algorithm based on HigherHRNet for the acquired hydraulic engineering field monitoring video images.
3. The engineering anomaly early warning system based on intelligent visual analysis as claimed in claim 2, wherein: the server is provided with server detection software for controlling the monitoring video information and the abnormal behavior detection result of the camera, displaying the result in a software interface and storing the result in a database, wherein the server detection software comprises a user management module, a data management module, an abnormal detection module and an alarm module; the abnormity detection module monitors a water conservancy building scene in real time and displays the detection condition in a display screen, and meanwhile, when abnormity occurs, the alarm module gives an alarm and captures the alarm and stores the captured image in a database folder.
4. The engineering anomaly early warning system based on intelligent visual analysis as claimed in claim 3, wherein: the abnormality detection module includes:
the life jacket wearing detection module is used for acquiring a hydraulic engineering field monitoring video image in real time, identifying a life jacket target by adopting a YOLOv4 algorithm, and judging whether a worker correctly wears the life jacket by combining a human posture identification algorithm based on HigherHRNet;
and the abnormal dangerous action detection module judges whether the worker is abnormal or not by the relative position relation between the target coordinate position detected in the image and the posture key point.
5. The engineering anomaly early warning system based on intelligent visual analysis as claimed in claim 4, wherein: the abnormal dangerous action detection module comprises:
the falling detection module and the climbing detection module judge whether the worker falls or not and whether climbing behavior occurs or not through the relative position relation between the target coordinate position detected in the image and the posture key point;
a danger area detection module; whether workers invade the dangerous area or not is judged by detecting the workers in the dangerous area marked in the hydraulic engineering area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110753861.7A CN113506416A (en) | 2021-07-03 | 2021-07-03 | Engineering abnormity early warning method and system based on intelligent visual analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110753861.7A CN113506416A (en) | 2021-07-03 | 2021-07-03 | Engineering abnormity early warning method and system based on intelligent visual analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113506416A true CN113506416A (en) | 2021-10-15 |
Family
ID=78011146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110753861.7A Pending CN113506416A (en) | 2021-07-03 | 2021-07-03 | Engineering abnormity early warning method and system based on intelligent visual analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113506416A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114449217A (en) * | 2021-12-24 | 2022-05-06 | 中国船舶重工集团公司第七一五研究所 | Equipment and method for dynamically monitoring video of shipborne retraction equipment based on machine vision |
| CN115600438A (en) * | 2022-11-25 | 2023-01-13 | 河海大学（Cn） | Data processing method and system suitable for hydraulic engineering |
| CN116437216A (en) * | 2023-06-12 | 2023-07-14 | 湖南博信创远信息科技有限公司 | Engineering supervision method and system based on artificial intelligence data processing and visual analysis |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107845239A (en) * | 2017-12-01 | 2018-03-27 | 同济大学 | A kind of operation on the sea personal security status monitoring and early warning system |
| CN109190569A (en) * | 2018-09-11 | 2019-01-11 | 锐马（福建）电气制造有限公司 | A kind of shared marine vehicle passenger life jacket wearing identifying system |
| CN110502965A (en) * | 2019-06-26 | 2019-11-26 | 哈尔滨工业大学 | A kind of construction safety helmet wearing monitoring method based on the estimation of computer vision human body attitude |
| CN112542027A (en) * | 2020-12-04 | 2021-03-23 | 国网浙江德清县供电有限公司 | Construction site personnel safety warning system based on image recognition |
| CN112651642A (en) * | 2020-12-29 | 2021-04-13 | 中水三立(北京)技术服务有限公司 | Hydraulic engineering field personnel safety management system based on multi-device perception |
| CN112819068A (en) * | 2021-01-29 | 2021-05-18 | 南京长江油运有限公司 | Deep learning-based real-time detection method for ship operation violation behaviors |
-
2021
- 2021-07-03 CN CN202110753861.7A patent/CN113506416A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107845239A (en) * | 2017-12-01 | 2018-03-27 | 同济大学 | A kind of operation on the sea personal security status monitoring and early warning system |
| CN109190569A (en) * | 2018-09-11 | 2019-01-11 | 锐马（福建）电气制造有限公司 | A kind of shared marine vehicle passenger life jacket wearing identifying system |
| CN110502965A (en) * | 2019-06-26 | 2019-11-26 | 哈尔滨工业大学 | A kind of construction safety helmet wearing monitoring method based on the estimation of computer vision human body attitude |
| CN112542027A (en) * | 2020-12-04 | 2021-03-23 | 国网浙江德清县供电有限公司 | Construction site personnel safety warning system based on image recognition |
| CN112651642A (en) * | 2020-12-29 | 2021-04-13 | 中水三立(北京)技术服务有限公司 | Hydraulic engineering field personnel safety management system based on multi-device perception |
| CN112819068A (en) * | 2021-01-29 | 2021-05-18 | 南京长江油运有限公司 | Deep learning-based real-time detection method for ship operation violation behaviors |
Non-Patent Citations (2)
| Title |
|---|
| 于乃功等: "基于姿态估计的实时跌倒检测算法", 《控制与决策》 * |
| 陈颖鸣等: "智能视频监控中异常行为识别研究", 《微电子学与计算机》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114449217A (en) * | 2021-12-24 | 2022-05-06 | 中国船舶重工集团公司第七一五研究所 | Equipment and method for dynamically monitoring video of shipborne retraction equipment based on machine vision |
| CN115600438A (en) * | 2022-11-25 | 2023-01-13 | 河海大学（Cn） | Data processing method and system suitable for hydraulic engineering |
| CN115600438B (en) * | 2022-11-25 | 2023-03-07 | 河海大学 | Data processing method and system suitable for hydraulic engineering |
| CN116437216A (en) * | 2023-06-12 | 2023-07-14 | 湖南博信创远信息科技有限公司 | Engineering supervision method and system based on artificial intelligence data processing and visual analysis |
| CN116437216B (en) * | 2023-06-12 | 2023-09-08 | 湖南博信创远信息科技有限公司 | Engineering supervision method and system based on artificial intelligence data processing and visual analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111144263B (en) | Construction worker high-falling accident early warning method and device | |
| CN113506416A (en) | Engineering abnormity early warning method and system based on intelligent visual analysis | |
| CN111445524B (en) | Scene understanding-based construction site worker unsafe behavior identification method | |
| Shrestha et al. | Hard‐hat detection for construction safety visualization | |
| CN110414400B (en) | Automatic detection method and system for wearing of safety helmet on construction site | |
| CN110795989A (en) | Intelligent safety monitoring system for electric power operation and monitoring method thereof | |
| CN106412501A (en) | Construction safety behavior intelligent monitoring system based on video and monitoring method thereof | |
| CN111341068B (en) | Construction method of drilling site dangerous area early warning system based on deep learning | |
| CN111428617A (en) | Video image-based distribution network violation maintenance behavior identification method and system | |
| CN111079722B (en) | Hoisting process personnel safety monitoring method and system | |
| CN114973140A (en) | Dangerous area personnel intrusion monitoring method and system based on machine vision | |
| CN112488483A (en) | AI technology-based EHS transparent management system and management method | |
| CN110121053A (en) | A kind of video monitoring method of situ of drilling well risk stratification early warning | |
| CN114821406A (en) | Method and system for judging electric power operation field violation behaviors | |
| CN112800901A (en) | Mine personnel safety detection method based on visual perception | |
| CN113206978A (en) | Security intelligent monitoring early warning system and method for oil and gas pipeline station | |
| Zhu et al. | Automated monitoring of hardhats wearing for onsite safety enhancement | |
| CN115100813B (en) | Intelligent community system based on digital twins | |
| CN108388873A (en) | A kind of supervisory systems of machine vision, method and client computer, storage medium | |
| CN105761423B (en) | Detect the device and method of the behavior of the user of hand-held electronic equipment | |
| CN116672629A (en) | Monitoring method and application of safety hook for overhead working personnel | |
| JP6977607B2 (en) | Safety judgment device, safety judgment system, safety judgment method | |
| CN113762115B (en) | Distribution network operator behavior detection method based on key point detection | |
| CN115953815A (en) | Monitoring method and device for infrastructure site | |
| CN114359831A (en) | Risk omen reasoning-oriented intelligent identification system and method for worker side-falling |
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: 20211015 |