CN116437216B - Engineering supervision method and system based on artificial intelligence data processing and visual analysis - Google Patents

Engineering supervision method and system based on artificial intelligence data processing and visual analysis Download PDF

Info

Publication number
CN116437216B
CN116437216B CN202310686725.XA CN202310686725A CN116437216B CN 116437216 B CN116437216 B CN 116437216B CN 202310686725 A CN202310686725 A CN 202310686725A CN 116437216 B CN116437216 B CN 116437216B
Authority
CN
China
Prior art keywords
target
operator
human body
ball machine
visual analysis
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202310686725.XA
Other languages
Chinese (zh)
Other versions
CN116437216A (en
Inventor
向帅
鲁靖
杨蜜
刘涛
周前峰
刘剑喜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan Bochuang High Tech Industrial Co ltd
Original Assignee
Hunan Boxin Chuangyuan Information Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hunan Boxin Chuangyuan Information Technology Co ltd filed Critical Hunan Boxin Chuangyuan Information Technology Co ltd
Priority to CN202310686725.XA priority Critical patent/CN116437216B/en
Publication of CN116437216A publication Critical patent/CN116437216A/en
Application granted granted Critical
Publication of CN116437216B publication Critical patent/CN116437216B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • H04N23/695Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01BMEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00Measuring arrangements characterised by the use of optical techniques
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01VGEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V8/00Prospecting or detecting by optical means
    • G01V8/10Detecting, e.g. by using light barriers
    • G01V8/20Detecting, e.g. by using light barriers using multiple transmitters or receivers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/90Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02PCLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30Computing systems specially adapted for manufacturing
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04SSYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geophysics (AREA)
  • Alarm Systems (AREA)

Abstract

The invention discloses an engineering supervision method and system based on artificial intelligence data processing and visual analysis, wherein the engineering supervision system comprises a plurality of ball machines and a plurality of human body infrared sensors, which are arranged in a monitoring range, wherein the ball machines and the human body infrared sensors are respectively in communication connection with edge proxy equipment, and the edge proxy equipment is used for being in communication connection with a cloud server; the method comprises the following steps: detecting the position of an operator through a human body infrared sensor, determining a ball machine for shooting the position of the operator by using an edge proxy device, controlling the shooting angle of the ball machine for shooting the position of the operator to be adjusted to face the position of the operator, and receiving an operation image acquired by the ball machine for shooting the position of the operator; the job image is used as input data to be input into a first visual analysis model of the edge proxy device. The invention is beneficial to automatically adjusting the direction of video monitoring and improving the identification effect of personnel which do not meet the operation requirement in the supervision process.

Description

Engineering supervision method and system based on artificial intelligence data processing and visual analysis
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to an engineering supervision method and an engineering supervision system based on artificial intelligence data processing and visual analysis.
Background
Engineering construction operations are often accompanied by dangerous factors, for example, building construction operations or electric power construction operations are usually located at high altitudes, the operation conditions and environments of which are more complex than those of ground operations, and data show that high altitude falling is the most serious accident in various safety accidents in the construction industry. The high-rise operation environment is complex and various, operators need to move frequently in the operation process, and in this case, it is very important to ensure the life safety of the operators. Therefore, it is required that the worker be able to meet the work requirements and wear the safety protection device correctly.
However, due to insufficient safety awareness of some operators, the situation that the operation requirements are not met or the safety protection equipment is not worn normally occurs, and a large potential safety hazard is caused for the construction operation.
Taking the example of wearing the safety belt by an operator, the safety belt worn by the operator at high altitude is usually provided with double safety hanging rings, 2 hanging rings are hung at the position above the waist of the operator during high altitude operation and rest, and at least 1 hanging ring is required to be safely and correctly worn during movement, so that the safety belt is beneficial to reducing the impact distance during accidental falling, and is a correct safety belt tying and hanging method. Although the safety belt is provided with the high-altitude operation protection equipment with good design at present, accidents are still occasional, and on one hand, the safety belt is difficult to ensure to play a real protection function only by means of conscious awareness of operators. For example, due to irregular wearing or untimely adjustment of the positions of the hooks, a part of operators have many incorrect wearing operations such as unhooking, low hanging and high use, and the like, and the buffer function of the safety belt cannot be effectively exerted when the accidental occurrence happens, so that the irrecoverable personal injury is caused. On the other hand, the actual operation process lacks an effective supervision mode because of limited manpower and equipment conditions, and the improper operation in the construction process of operators cannot be timely reminded.
In the engineering supervision method in the prior art, the operation process is supervised manually or by adopting a monitoring video, the reliability of the manual supervision mode is not high, the monitoring video is always fixed-point monitoring, a person which does not meet the operation requirement is required to be searched for by a monitoring person through a video picture obtained by fixed-point monitoring, so that the behavior of dangerous operation is difficult to capture, and the dangerous operation person is difficult to effectively remind, therefore, the engineering supervision method and the system based on artificial intelligent data processing and visual analysis are needed to be provided, the direction of the video monitoring is automatically adjusted, and the identification effect of the person which does not meet the operation requirement in the engineering supervision process is improved.
Disclosure of Invention
The invention mainly aims to provide an engineering supervision method based on artificial intelligence data processing and visual analysis, which aims to automatically adjust the direction of video monitoring and improve the identification effect of personnel which do not meet the operation requirements in the engineering supervision process.
In order to achieve the above purpose, the engineering supervision method based on artificial intelligence data processing and visual analysis is applied to an engineering supervision system, wherein the engineering supervision system comprises a plurality of ball machines and a plurality of human body infrared sensors, which are arranged in a monitoring range, the ball machines and the human body infrared sensors are respectively in communication connection with edge proxy equipment, and the edge proxy equipment is used for being in communication connection with a cloud server; the method comprises the following steps:
Detecting the position of an operator through a human body infrared sensor, determining a ball machine for shooting the position of the operator by using an edge proxy device, controlling the shooting angle of the ball machine for shooting the position of the operator to be adjusted to face the position of the operator, and receiving an operation image acquired by the ball machine for shooting the position of the operator;
and taking the operation image as input data, and inputting the operation image into a first visual analysis model of the edge proxy equipment.
Preferably, the method further comprises:
the first visual analysis model identifies target operators which do not meet the operation requirements from the operation image, and outputs the target operators and target operation positions corresponding to the target operators;
according to the target operation position, determining a target ball machine for monitoring target operators and sending out an alarm prompt;
the target ball machine is controlled to continuously collect real-time operation pictures of target operators within a set time length, and the real-time operation pictures are sent to the edge proxy equipment;
judging whether the edge proxy equipment establishes communication connection with a cloud server or not;
when the edge proxy equipment does not establish communication connection with the cloud server, the edge proxy equipment inputs the acquired multiple real-time operation pictures into the first visual analysis model again as input data;
When the first visual analysis model outputs the target operator and the target operation position which do not meet the operation requirement again, determining the target operator and the target operation position as temporary analysis results, and sending the temporary analysis results to the management terminal;
and when the output result of the first visual analysis model is empty, releasing the continuous acquisition instruction of the target ball machine to the target operator.
Preferably, after the step of determining whether the edge proxy device establishes a communication connection with the cloud server, the method further includes:
when the edge proxy equipment and the cloud server are in communication connection, the edge proxy equipment sends a plurality of acquired real-time operation pictures to the cloud server;
the cloud server inputs each real-time operation picture as input data into a second visual analysis model;
when the second visual analysis model outputs target operators and target operation positions which do not meet the operation requirements, determining the target operators and the target operation positions as formal analysis results, and sending the formal analysis results to the management terminal;
and when the output result of the second visual analysis model is empty, releasing the continuous acquisition instruction of the target ball machine to the target operator.
Preferably, the step of determining a target ball machine for monitoring the target operator according to the target work position includes:
determining a ball set for enclosing the target operation position according to the target operation position and the installation position of each ball machine, wherein the ball set comprises a plurality of ball machines which jointly encircle the target operation position;
and controlling each ball machine of the ball machine set to adjust the shooting angle to face the target operator according to the target operation position so as to form the ball machine set for monitoring the target operator.
Preferably, the step of continuously collecting real-time operation pictures of target operators and sending the real-time operation pictures to the edge proxy device by the control target ball machine in a set time length comprises the following steps:
forming a multi-angle monitoring image of a target operator through real-time operation images respectively acquired by each ball machine in the ball machine set;
the edge proxy equipment sends the multi-angle monitoring images at the same acquisition time point to different layers according to different visual angles;
scaling deformation is carried out on images with different visual angles in each image layer so as to fuse multi-angle monitoring images with the same acquisition time point to form a three-dimensional operation picture with the same acquisition time point.
Preferably, the method further comprises:
Establishing an image library storing a plurality of target images, wherein the target images comprise multi-angle images in a standard working state;
simplifying the multi-angle image according to the standard operation state at the cloud server to obtain an abstract graph;
the abstract graph is sent to the first visual analysis model and the second visual analysis model to identify a standard job status from the input image by the first visual analysis model and the second visual analysis model.
Preferably, the edge proxy device sets up a processing module; the method further comprises the steps of:
acquiring the position of an operator detected by a human body infrared sensor and a shooting period corresponding to the position of each operator by a ball machine so as to form first monitoring data;
the processing module forms operation dynamic record data of operators according to the first monitoring data;
acquiring each target operator, a target operation position corresponding to each target operator and a corresponding shooting period to form second monitoring data;
and the processing module forms nonstandard operation data according to the second monitoring data, wherein the nonstandard operation data comprises personnel, operation positions and operation time periods which do not meet the operation requirements.
Preferably, the method further comprises:
after monitoring starts, the edge proxy equipment determines the start and end of the starting time of the operation picture according to the infrared data collected by the human body infrared sensor and the image collected by the dome camera;
the edge proxy equipment marks the time zone which falls into the operation picture starting time and the time zone which does not fall into the operation picture starting time respectively;
and storing the pictures acquired by the ball machine in the time section of the operation picture starting time and the pictures acquired by the ball machine in the time section of the operation picture starting time to different storage positions of the edge proxy equipment.
Preferably, human infrared sensors are arranged at different positions in the monitoring range, the positions of operators are detected through the human infrared sensors, and the edge proxy equipment determines a ball machine for shooting the positions of the operators, and the method comprises the following steps:
each human body infrared sensor senses whether a human body enters a detection range or not according to the same detection period;
when a human body infrared sensor for entering a detection range exists, the corresponding detection range is used as a fuzzy position area;
judging whether each fuzzy position area has intersection with other fuzzy position areas or not;
When no intersection exists, determining the position of the operator according to each fuzzy position area, and determining a ball machine for shooting the position of the operator according to each fuzzy position area;
when an intersection exists, determining an accurate position area according to the intersection area of each fuzzy position area and other fuzzy position areas, and determining a ball machine for shooting the position of an operator according to the accurate position area; wherein the precise location area is the center of the intersection area;
wherein ,for the f-th ball machine, < >>F is the number of ball machines; />For the set of ball machines corresponding to the ith operator position, < >>I is the number of operators; />In order to detect the fuzzy position area corresponding to the human infrared sensor with the serial number k of the ith operator,/for the human infrared sensor>Is a fuzzy position area corresponding to the human body infrared sensor with the serial number of j, and is +.>N is the number of human body infrared sensors;
when detecting that the center of the operator position corresponding to the human body infrared sensor with the number k of the ith operator is the center of the fuzzy position area corresponding to the human body infrared sensor with the number k, the user is at the moment->The distance from the f-th dome camera to the center of a fuzzy position area corresponding to the human body infrared sensor with the serial number k;
When the center of the position of the operator corresponding to the human body infrared sensor with the serial number k of the ith operator is detected to be the center of the precise position area corresponding to the human body infrared sensor with the serial number k, at the moment,the distance from the f-th ball machine to the center of the accurate position area corresponding to the human body infrared sensor with the serial number k;
the distance from the g-th dome camera to the center of the fuzzy position area corresponding to the human infrared sensor with the serial number k is->For the distance from the g-th ball machine to the center of the precise position area corresponding to the human body infrared sensor with the serial number k,
the invention also provides an engineering supervision system, which is applied to the engineering supervision method based on the artificial intelligent data processing and the visual analysis, and comprises a plurality of ball machines and a plurality of human body infrared sensors which are arranged in a monitoring range, wherein the ball machines and the human body infrared sensors are respectively in communication connection with edge proxy equipment, and the edge proxy equipment is used for being in communication connection with a cloud server.
The technical scheme of the invention has the following beneficial effects:
the intelligent monitoring system comprises a plurality of ball machines and a plurality of human body infrared sensors, wherein the ball machines and the human body infrared sensors are arranged in a monitoring range and are respectively in communication connection with edge proxy equipment, the human body infrared sensors are used for detecting whether an operator enters a detection range through an infrared detection principle, and accordingly the edge proxy equipment can determine the approximate operation position of the operator according to infrared data detected by the human body infrared sensors. After the work position is determined, the edge proxy device determines a ball machine for shooting the position of the work personnel from the ball machines, and controls the shooting angle of the ball machine for shooting the position of the work personnel to be adjusted towards the position of the work personnel. Therefore, each ball machine in the invention does not randomly collect images in a monitoring range, but collects operation images according to a designated direction, namely according to the operation position of an operator, and sends the operation images to the edge proxy equipment for visual analysis. Thus, the edge proxy device receives all job status images that have visual analysis significance, rather than invalid images where no job personnel are present. Therefore, the invention can avoid the defect that in the prior art, operators are not standardized to be searched from the monitoring picture by monitoring personnel, so that the invention is beneficial to automatically adjusting the direction of video monitoring and improving the identification effect of the operators which do not meet the operation requirement in the engineering supervision process.
And after the human body infrared sensor is positioned at the position of the operator, the ball machine performs targeted operation image acquisition on the operator so as to find out the target operator which does not meet the operation requirement through a visual analysis method from the acquired image, and then the target operator is directly positioned through the ball machine, so that one or more ball machines can perform special image acquisition on the target operator in the process, and the special image acquisition on the target operator which does not meet the operation requirement is facilitated.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an embodiment of an engineering supervision method based on artificial intelligence data processing and visual analysis according to the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, in order to achieve the above objective, a first embodiment of the present invention proposes an engineering supervision method based on artificial intelligence data processing and visual analysis, where the engineering supervision system includes a plurality of ball machines and a plurality of human body infrared sensors disposed in a monitoring range, the ball machines and the human body infrared sensors are respectively in communication connection with an edge proxy device, and the edge proxy device is used for communication connection with a cloud server; the method comprises the following steps:
step S10, detecting the position of an operator through a human body infrared sensor, determining a ball machine for shooting the position of the operator by using an edge proxy device, controlling the shooting angle of the ball machine for shooting the position of the operator to be adjusted to face the position of the operator, and receiving an operation image acquired by the ball machine for shooting the position of the operator;
and step S20, using the operation image as input data, and inputting a first visual analysis model of the edge proxy device.
The intelligent monitoring system comprises a plurality of ball machines and a plurality of human body infrared sensors, wherein the ball machines and the human body infrared sensors are arranged in a monitoring range and are respectively in communication connection with edge proxy equipment, the human body infrared sensors are used for detecting whether an operator enters a detection range through an infrared detection principle, and accordingly the edge proxy equipment can determine the approximate operation position of the operator according to infrared data detected by the human body infrared sensors. After the work position is determined, the edge proxy device determines a ball machine for shooting the position of the work personnel from the ball machines, and controls the shooting angle of the ball machine for shooting the position of the work personnel to be adjusted towards the position of the work personnel. Therefore, each ball machine in the invention does not randomly collect images in a monitoring range, but collects operation images according to a designated direction, namely according to the operation position of an operator, and sends the operation images to the edge proxy equipment for visual analysis. Thus, the edge proxy device receives all job status images that have visual analysis significance, rather than invalid images where no job personnel are present. Therefore, the invention can avoid the defect that in the prior art, operators are not standardized to be searched from the monitoring picture by monitoring personnel, so that the invention is beneficial to automatically adjusting the direction of video monitoring and improving the identification effect of the operators which do not meet the operation requirement in the engineering supervision process.
And after the human body infrared sensor is positioned at the position of the operator, the ball machine performs targeted operation image acquisition on the operator so as to find out the target operator which does not meet the operation requirement through a visual analysis method from the acquired image, and then the target operator is directly positioned through the ball machine, so that one or more ball machines can perform special image acquisition on the target operator in the process, and the special image acquisition on the target operator which does not meet the operation requirement is facilitated.
Further, although the edge proxy device can directly communicate with the on-site dome camera and the human body infrared sensor, the amount of monitoring data generated by the dome camera and the human body infrared sensor in real time is large, the edge proxy device also needs to store a part of historical data, and perform data processing and visual analysis on the real-time data, and the processing capacity and the storage space of the edge proxy device are limited, so that the collection of invalid monitoring data of no-operation personnel is avoided through the combination of the human body infrared sensor and the dome camera, the total data collection amount is reduced, real-time engineering supervision is realized through the limited storage space and the limited processing capacity of the edge proxy device, and the dependence degree of data processing and visual analysis which are required to be performed by a cloud server is reduced. The control of the shooting angle of the dome camera is beneficial to improving the purpose of acquiring images by the dome camera, so that the images input into the edge proxy equipment and the images needing to be subjected to visual analysis in the edge proxy equipment are all images containing operators and are not invalid images needing to be removed, thereby being beneficial to reducing the data quantity stored and processed by the edge proxy equipment and improving the visual analysis accuracy of the edge proxy equipment.
In addition, the edge proxy equipment is arranged, and because good network signals are not usually available nearby the construction engineering, data respectively collected by the dome camera and the human body infrared sensor cannot be necessarily sent to the cloud server at any time for real-time visual analysis and data processing. However, the edge proxy device can be arranged in the monitoring range and can be directly communicated with the ball machine and the human body infrared sensor in the monitoring range, so that the edge proxy device can store data acquired by the ball machine and the human body infrared sensor in real time and can process the data acquired by the ball machine and the human body infrared sensor in real time and perform visual analysis. And engineering supervision can be performed in real time under the condition of poor network signals.
The edge proxy equipment is used as edge computing equipment, has the characteristics of compatibility equipment, cloud service, cloud platform collaborative management, intelligent algorithm maintenance, data storage and the like, and is suitable for monitoring various monitoring scenes, such as intelligent edge scene monitoring of electric pole installation, transformer substations, indoor station rooms, intelligent security, wind power plants, intelligent factories, intelligent rail transit, internet of vehicles, intelligent oil gas and the like.
The edge proxy device has unified proxy service of data message and standardization of communication protocol, and supports data encryption transmission; the complex edge calculation is supported, and micro-applications such as data analysis, intelligent video processing and the like can be flexibly configured; the method can realize the deployment, upgrading and operation management of the edge algorithm micro-application and software based on the container technology; meanwhile, the system has the functions of data storage, data playback, deep analysis and the like.
The ball machine and the human body infrared sensor are respectively used as data acquisition terminals in communication connection with the edge proxy equipment.
The edge proxy equipment comprises a data storage module, a processing module and a visual analysis module, wherein the image acquired by the dome camera and the infrared data acquired by the human body infrared sensor, and the data processing results of the processing module and the visual analysis module are stored in the data storage module. The processing module is used for performing data processing outside visual analysis on the image acquired by the ball machine and the infrared data acquired by the human body infrared sensor, and the visual analysis module is used for operating the first visual analysis model to perform visual analysis on the image.
Under the condition of good network state, the edge proxy equipment can establish communication connection with the cloud server, the final result of visual analysis is obtained by using the computing resources of the cloud server, and the data stored in the data storage module of the edge proxy equipment are uploaded to the cloud server for cloud backup storage, so that the data already backed up by the cloud can be deleted, and the storage space is released.
According to the invention, each ball machine can realize multi-angle rotation, and the rotation angle range can be set according to requirements, for example, the rotation angle is 0-355 degrees.
Based on the first embodiment of the present invention, in a second embodiment of the engineering supervision method based on artificial intelligence data processing and visual analysis of the present invention, the method further includes:
step S30, a first visual analysis model identifies target operators which do not meet the operation requirements from the operation images, and outputs the target operators and target operation positions corresponding to the target operators;
step S40, determining a target ball machine for monitoring target operators according to the target operation position, and sending out an alarm prompt;
step S50, the target ball machine is controlled to continuously collect real-time operation pictures of target operators within a set time length, and the real-time operation pictures are sent to the edge proxy equipment;
step S60, judging whether the edge proxy equipment establishes communication connection with a cloud server;
step S70, when the edge proxy equipment does not establish communication connection with the cloud server, the edge proxy equipment inputs the acquired multiple real-time operation pictures into the first visual analysis model again as input data;
step S80, when the first visual analysis model outputs the target operator and the target operation position which do not meet the operation requirement again, determining the target operator and the target operation position as temporary analysis results, and sending the temporary analysis results to the management terminal;
And step S90, when the output result of the first visual analysis model is empty, releasing the continuous acquisition instruction of the target ball machine to the target operator.
The invention meets the operation requirement, namely the operation meets the operation requirement and/or safety protection equipment is worn in the operation process in a standardized way, and the safety protection equipment can be safety belts, safety helmets, protective masks, gloves, insulating shoes, protective glasses and the like, and can also be other safety protection equipment, but is not limited to the safety protection equipment. By non-conforming to the operational requirements in the present invention is meant that the operational operations do not conform to the operational requirements and/or that safety protection equipment is not being worn properly during the operation.
The real-time operation picture collected by the target operator is subjected to visual analysis, so that whether the target operator does not meet the operation requirement or not is confirmed through special shooting and visual analysis of the target operator.
If the first visual analysis model outputs the target operator and the target work position again (which may be the same or different from the target operator and the target work position output for the first time), a temporary analysis result is formed. The temporary analysis result can be received through the management terminal, and further, the alarm prompt can be carried out through the alarm device in the monitoring range.
If the output result of the first visual analysis model is empty, the fact that the collected images are monitored in a centralized mode through all the ball machines does not exist behavior which does not meet the operation requirement is indicated, and the continuous collection instruction of the target ball machine can be released.
In order to save the computing resources and the storage space of the edge proxy equipment, the connection period of the edge proxy equipment actively connected with the cloud server can be set, so that the edge proxy equipment can actively request to establish communication connection with the cloud server, and the automatic connection between the edge proxy equipment and the cloud server is realized. Furthermore, when detecting a connection signal sent by a user, the edge proxy device and the cloud server can be connected in a communication mode.
The processing module of the edge proxy device is used for directly receiving external data and processing the data according to the data type as follows:
when the processing module receives infrared data sent by the human infrared sensor, the infrared data sent by the human infrared sensor are stored in a first storage space of the data storage module according to the time stamp, and the infrared data sent by the human infrared sensor are analyzed to determine the position of an operator;
when the processing module receives the image sent by the dome camera, the image sent by the dome camera is stored in a second storage space of the data storage module according to the time stamp of the data, and the image required to be sent to the first visual analysis model is preprocessed, wherein at least one of noise in the image is removed, smoothing is performed, contrast is enhanced, and color balance is adjusted. The processing module sends the preprocessed data to a first visual analysis model, and the first visual analysis model extracts a portrait from the operation image so as to identify a corresponding target operation person and judges whether a characteristic image meeting the operation requirement can be extracted from the operation image.
If not, the operator does not meet the operation requirement, and outputting the target operator which does not meet the operation requirement from the first visual analysis model. Meanwhile, since the operation image is acquired after the human body infrared sensor detects the position of the operator, the position of the operator detected by the human body infrared sensor can be used as a target operation position and also output from the first visual analysis model.
If so, the operator meets the operation requirement, and the output result of the first visual analysis model is null.
In a third embodiment of the engineering supervision method based on artificial intelligence data processing and visual analysis according to the second embodiment of the present invention, after the step S60, the method further includes:
step S100, when the edge proxy equipment and the cloud server are in communication connection, the edge proxy equipment sends a plurality of acquired real-time operation pictures to the cloud server;
step S110, the cloud server inputs each real-time operation picture as input data into a second visual analysis model;
step S120, when the second visual analysis model outputs the target operator and the target operation position which do not meet the operation requirement, determining the target operator and the target operation position as formal analysis results, and transmitting the formal analysis results to the management terminal;
And step S130, when the output result of the second visual analysis model is empty, releasing the continuous acquisition instruction of the target ball machine to the target operator.
The edge proxy equipment detects whether communication connection is established with the cloud server or not, and detects whether communication between the edge proxy equipment and the cloud server is interrupted or not after the communication connection is established. If so, steps S70 to S90 are performed.
The second embodiment and the third embodiment may be used in superposition. Specifically, when the edge proxy device and the cloud server are not in communication connection, the second embodiment may be used to perform visual analysis first, and until the edge proxy device and the cloud server are in communication connection, the third embodiment is used to perform visual analysis.
In a fourth embodiment of the engineering supervision method based on artificial intelligence data processing and visual analysis according to the second to third embodiments of the present invention, the step of determining the target ball machine for monitoring the target operator according to the target operation position in the step S40 includes:
step S41, determining a ball set for enclosing a target operation position according to the target operation position and the installation position of each ball machine, wherein the ball set comprises a plurality of ball machines which jointly encircle the target operation position;
In step S42, each ball machine of the ball machine set is controlled to adjust the shooting angle to face the target operator according to the target operation position, so as to form the ball machine set for monitoring the target operator.
Specifically, there are a plurality of ball machines within the monitoring range. After the positions of the target operators which do not meet the operation requirements are determined, the target ball machine for monitoring the target operators is determined, so that the target operators can be specially monitored for a long time through the target ball machine, real-time operation pictures for a period of time are specially acquired for the target operators, and the real-time operation pictures are sent to the first visual analysis model for visual analysis.
The target ball machine can comprise a special ball machine or a ball machine set comprising a plurality of ball machines.
When the target ball machine is a special ball machine, the ball machine is the ball machine closest to the target working position, so that the acquired image is more accurate and clear.
When the target ball machine is a ball machine set comprising a plurality of ball machines, the ball machine set is a plurality of ball machines which jointly encircle the target operation position so as to form multi-angle encircling shooting for target operators.
The position of each operating personnel in the monitoring range may be dispersed in different areas, at this time, there are a plurality of operating personnel positions, just need divide into a plurality of ball sets with the ball machine, divide into a plurality of ball sets's mode:
Firstly, determining the approximate position range of the position of an operator according to the existence of a human body in a detection area of a human body infrared sensor, then determining the distance between each ball machine and the boundary of the approximate position range, and dividing the ball machines with the distance smaller than a preset value into the same ball machine set.
In a fifth embodiment of the engineering supervision method according to the third or fourth embodiment of the present invention, the step S50 includes:
step S51, forming a multi-angle monitoring image of a target operator through real-time operation images respectively acquired by all ball machines in the ball machine set;
step S52, the edge proxy equipment sends the multi-angle monitoring images of the same acquisition time point to different layers according to different visual angles;
and step S53, scaling deformation is carried out on the images with different visual angles in each image layer so as to fuse the multi-angle monitoring images with the same acquisition time point to form a three-dimensional operation picture with the same acquisition time point.
Specifically, in the same ball machine set, each ball machine forms a circle around the periphery of the position of an operator, each ball machine is sequentially connected in series according to a anticlockwise or clockwise circle sequence, the serial numbers of each ball machine in the same ball machine set are formed according to the serial numbers of each ball machine, the serial numbers of images shot by each ball machine at the same acquisition time point are determined according to the serial numbers of each ball machine, and the positions of the images shot by each ball machine at the same acquisition time point are arranged according to the serial numbers of the images so as to realize multi-angle monitoring images.
In a sixth embodiment of the engineering supervision method according to the present invention based on artificial intelligence data processing and visual analysis, the method further includes:
step S140, an image library storing a plurality of target images is established, wherein the target images comprise multi-angle images in a standard operation state;
step S150, simplifying the multi-angle image according to the standard operation state at the cloud server to obtain an abstract figure;
step S160, transmitting the abstract graph to the first visual analysis model and the second visual analysis model to identify the standard operation state from the input image through the first visual analysis model and the second visual analysis model.
The standard operation state refers to a state meeting the operation requirement. The target images in the image library can be added or deleted at different time points according to the requirements, and after the target images are added or deleted, the analysis comparison standards of the first visual analysis model and the second visual analysis model can be correspondingly adjusted, so that the method can adapt to the change of the supervision requirements.
In a seventh embodiment of the engineering supervision method based on artificial intelligence data processing and visual analysis according to the present invention, the edge proxy device sets a processing module; the method further comprises the steps of:
Step S170, acquiring the position of an operator detected by a human body infrared sensor and a shooting period corresponding to the position of each operator by the ball machine so as to form first monitoring data;
step S180, the processing module forms operation dynamic record data of an operator according to the first monitoring data;
step S190, obtaining each target operator, a target operation position corresponding to each target operator and a corresponding shooting period to form second monitoring data;
in step S200, the processing module forms non-standard job data according to the second monitoring data, where the non-standard job data includes personnel, job positions and job periods that do not meet the job requirements.
The change of the working position of the working personnel with time can be determined according to the first monitoring data. According to the second monitoring data, the change of the working position of the person which does not meet the working requirements and the period of playing chess which does not meet the working requirements can be determined, so that the fault tracing or accident tracing in the engineering supervision process is facilitated.
Based on the seventh embodiment of the present invention, in an eighth embodiment of the engineering supervision method based on artificial intelligence data processing and visual analysis of the present invention, the method further includes:
Step S210, after monitoring starts, the edge proxy equipment determines the start and end of the starting time of the operation picture according to the infrared data collected by the human body infrared sensor and the image collected by the dome camera;
step S220, the edge proxy device marks the time zone which falls into the operation picture starting time and the time zone which does not fall into the operation picture starting time respectively;
and step S230, storing the pictures acquired by the dome camera in the time section of the operation picture starting time and the pictures acquired by the dome camera in the time section of the operation picture starting time to different storage positions of the edge proxy equipment.
Specifically, after monitoring starts, the edge proxy device controls the ball machine and the human body infrared sensor to start, the human body infrared sensor collects infrared data according to a preset detection period and sends the infrared data to the processing module of the edge proxy device, so that the processing module analyzes the operation position of an operator from the infrared data. Meanwhile, the edge proxy equipment also controls the ball machine to monitor the monitoring range according to a preset mode. Wherein, the infrared image and the operation image are marked with time stamps according to the acquisition time.
It is easy to understand that after the human body infrared sensor and the ball machine are started, an operator does not necessarily exist in the monitoring range at any time (for example, the operator leaves the working range in a working rest period, but the monitoring function is still needed, but the image acquired at this time does not need to identify whether the operator does not meet the working requirement or not). Therefore, the processing module receives the data of the human body infrared sensor and judges whether a human body exists in the monitoring range; when the human body does not exist, the processing module controls the image collected by the ball machine to be stored in the second storage space of the data storage module, and the image collected by the ball machine does not need to be sent to the first visual analysis model; when a human body exists, the processing module takes the acquisition time stamp of the infrared data of the detected human body as the operation picture starting time of the ball machine, and the processing module determines the position of the human body according to the infrared data of the human body infrared sensor, and adjusts the shooting angle of the ball machine to the position of the human body.
The processing module sends the image acquired after the ball machine is adjusted to the human body position to a first visual analysis model, and the first visual analysis model compares the acquired image of the person with the stored image of the operator to determine whether the person is the operator; if not, the processing module deletes the recorded operation picture starting time; if yes, the first visual analysis module identifies target operators which do not meet the operation requirements from the images, and outputs the target operators and target operation positions corresponding to the target operators. In the invention, the images acquired in the time when the ball machine falls into the operation picture are stored in a partitioning manner with the images acquired in the time when the ball machine does not fall into the operation picture. For example, an image acquired during the time when the ball machine falls within the operation screen is stored in a first subspace of the second storage space, and an image acquired during the time when the ball machine does not fall within the operation screen is stored in a second subspace of the second storage space. The partition storage has the advantages that the images in the working process can be independently extracted to be used as the operation monitoring images to be independently recorded, personnel which do not meet the operation requirements are identified in the construction process, the images in the non-working process are also independently stored to form other auxiliary monitoring records, and the data processing capacity of the first visual analysis model can be reduced.
Further, after the operation picture starts to be calculated, the human body infrared sensor continuously monitors according to the set detection period, and as long as the human body infrared sensor detects a human body, the ball machine can locate a position corresponding to the human body and can send the acquired image to the first visual analysis model for visual analysis. And the operation picture starting time is ended until the image acquired after the ball machine is adjusted to the position facing the human body can not identify an operator, or the infrared image acquired by the human body infrared sensor does not find a character image any more.
Further, in the processing module, a goal machine picture recording table is formed, and in the goal machine picture recording table, according to the duration of the starting time of the goal machine, a time section falling into the starting time of the operation picture and a time section not falling into the starting time of the operation picture are respectively marked so as to automatically form monitoring time periods of two purposes. The application can also be used for automatically recording the working time of the operator.
Specifically, because the storage space of the edge proxy device is limited, if the edge proxy device does not establish communication connection with the cloud server for a period of time exceeding a set period of time, the data stored in the edge proxy device is not uploaded to the cloud server for a long time to be backed up, so that the data stored in the edge proxy device is more and more, and the remaining storage space of the edge proxy device is smaller than the set storage space. At this time, in order to enable the edge proxy device to still provide enough storage space to realize the monitoring function, the image acquired in the starting time of the operation picture may be deleted (i.e. the data stored in the second subspace of the second storage space is deleted), so as to increase the remaining available space of the edge proxy device, but not delete the effective operation monitoring data.
According to a seventh embodiment of the present invention, in a ninth embodiment of the engineering supervision method based on artificial intelligence data processing and visual analysis of the present invention, human infrared sensors are disposed at different positions within a monitoring range, and in the step S10, the step of detecting the position of the operator through the human infrared sensors, and the step of determining, by the edge proxy device, a dome camera for capturing the position of the operator includes:
step S11, each human body infrared sensor senses whether a human body enters a detection range or not according to the same detection period;
step S12, when a human body infrared sensor for entering a detection range exists, the corresponding detection range is used as a fuzzy position area;
step S13, judging whether each fuzzy position area has intersection with other fuzzy position areas;
step S14, when no intersection exists, determining the position of the operator according to each fuzzy position area, and determining a ball machine for shooting the position of the operator according to each fuzzy position area;
step S15, when an intersection exists, determining an accurate position area according to the intersection area of each fuzzy position area and other fuzzy position areas, and determining a dome camera for shooting the position of an operator according to the accurate position area; wherein the precise location area is the center of the intersection area;
wherein ,for the f-th ball machine, < >>F is the number of ball machines; />For the set of ball machines corresponding to the position of the ith operator (i.e. the ball set corresponding to the ith operator), the ball set is provided with a plurality of ball machines corresponding to the position of the ith operator>I is the number of operators; />In order to detect the fuzzy position area corresponding to the human infrared sensor with the serial number k of the ith operator,/for the human infrared sensor>Is a fuzzy position area corresponding to the human body infrared sensor with the serial number j,n is the number of human body infrared sensors;
when detecting that the center of the operator position corresponding to the human body infrared sensor with the number k of the ith operator is the center of the fuzzy position area corresponding to the human body infrared sensor with the number k, the user is at the moment->The distance from the f-th dome camera to the center of a fuzzy position area corresponding to the human body infrared sensor with the serial number k;
when the center of the position of the operator corresponding to the human body infrared sensor with the serial number k of the ith operator is detected to be the center of the precise position area corresponding to the human body infrared sensor with the serial number k, at the moment,the distance from the f-th ball machine to the center of the accurate position area corresponding to the human body infrared sensor with the serial number k;
the distance from the g-th dome camera to the center of the fuzzy position area corresponding to the human infrared sensor with the serial number k is- >For the distance from the g-th ball machine to the center of the precise position area corresponding to the human body infrared sensor with the serial number k,
therefore, the ball machines for photographing the positions of the operators are mainly determined by the distance between each ball machine and the position of the operator.
In the invention, the position of the operator is detected by the human body infrared sensor, and the edge proxy equipment determines the ball machine for shooting the position of the operator, so that a mapping relation table is established in the edge proxy equipment, and the corresponding relation between the position of the operator and the ball machine for shooting the position of the operator is shown by the mapping relation table.
Therefore, after the target operator which does not meet the operation requirement is identified through the operation image, the target operation position corresponding to the target operator is determined by inquiring the mapping relation table according to the ball machine which shoots the operation image.
In addition, in order to achieve the above purpose, the invention also provides an engineering supervision system, which applies the engineering supervision method based on artificial intelligence data processing and visual analysis, wherein the engineering supervision system comprises a plurality of ball machines and a plurality of human body infrared sensors which are arranged in a monitoring range, the ball machines and the human body infrared sensors are respectively in communication connection with edge proxy equipment, and the edge proxy equipment is used for being in communication connection with a cloud server.
The foregoing description of the preferred embodiments of the present invention should not be construed as limiting the scope of the invention, but rather utilizing equivalent structural changes made in the present invention description and drawings or directly/indirectly applied to other related technical fields are included in the scope of the present invention.

Claims (8)

1. The engineering supervision method based on the artificial intelligence data processing and visual analysis is characterized by being applied to an engineering supervision system, wherein the engineering supervision system comprises a plurality of ball machines and a plurality of human body infrared sensors, wherein the ball machines and the human body infrared sensors are arranged in a monitoring range and are respectively in communication connection with edge proxy equipment, and the edge proxy equipment is used for being in communication connection with a cloud server; human body infrared sensors are arranged at different positions in the monitoring range; the method comprises the following steps:
detecting the position of an operator through a human body infrared sensor, determining a ball machine for shooting the position of the operator by using an edge proxy device, controlling the shooting angle of the ball machine for shooting the position of the operator to be adjusted to face the position of the operator, and receiving an operation image acquired by the ball machine for shooting the position of the operator;
Inputting a first visual analysis model of the edge proxy device by taking the operation image as input data;
the first visual analysis model identifies target operators which do not meet the operation requirements from the operation image, and outputs the target operators and target operation positions corresponding to the target operators;
according to the target operation position, determining a target ball machine for monitoring target operators and sending out an alarm prompt;
the target ball machine is controlled to continuously collect real-time operation pictures of target operators within a set time length, and the real-time operation pictures are sent to the edge proxy equipment;
judging whether the edge proxy equipment establishes communication connection with a cloud server or not;
when the edge proxy equipment does not establish communication connection with the cloud server, the edge proxy equipment inputs the acquired multiple real-time operation pictures into the first visual analysis model again as input data;
when the first visual analysis model outputs the target operator and the target operation position which do not meet the operation requirement again, determining the target operator and the target operation position as temporary analysis results, and sending the temporary analysis results to the management terminal;
when the output result of the first visual analysis model is empty, releasing a continuous acquisition instruction of the target ball machine to a target operator;
The step of detecting the position of the operator through the human body infrared sensor, and determining the ball machine for shooting the position of the operator by the edge proxy equipment comprises the following steps:
each human body infrared sensor senses whether a human body enters a detection range or not according to the same detection period;
when a human body infrared sensor for entering a detection range exists, the corresponding detection range is used as a fuzzy position area;
judging whether each fuzzy position area has intersection with other fuzzy position areas or not;
when no intersection exists, determining the position of the operator according to each fuzzy position area, and determining a ball machine for shooting the position of the operator according to each fuzzy position area;
when an intersection exists, determining an accurate position area according to the intersection area of each fuzzy position area and other fuzzy position areas, and determining a ball machine for shooting the position of an operator according to the accurate position area; wherein the precise location area is the center of the intersection area;
wherein ,for the f-th ball machine, < >>F is the number of ball machines; />For the set of ball machines corresponding to the ith operator position, < >>I is the number of operators; />In order to detect the fuzzy position area corresponding to the human infrared sensor with the serial number k of the ith operator,/for the human infrared sensor >Is a fuzzy position area corresponding to the human body infrared sensor with the serial number of j, and is +.>,/>,/>N is the number of human body infrared sensors;
when detecting that the center of the operator position corresponding to the human body infrared sensor with the number k of the ith operator is the center of the fuzzy position area corresponding to the human body infrared sensor with the number k, the user is at the moment->The distance from the f-th dome camera to the center of a fuzzy position area corresponding to the human body infrared sensor with the serial number k;
when the center of the operator position corresponding to the human body infrared sensor with the number k of the ith operator is detected as the center of the precise position area corresponding to the human body infrared sensor with the number k, at the moment, the user is added with the sensor>The distance from the f-th ball machine to the center of the accurate position area corresponding to the human body infrared sensor with the serial number k;
for the g-th ball machine to the serial numberk, distance of center of fuzzy position area corresponding to human body infrared sensor, +.>For the distance from the g-th dome camera to the center of the precise position area corresponding to the human infrared sensor with the serial number k,/for the human infrared sensor>
2. The engineering supervision method based on artificial intelligence data processing and visual analysis according to claim 1, further comprising, after the step of determining whether the edge proxy device establishes a communication connection with the cloud server:
When the edge proxy equipment and the cloud server are in communication connection, the edge proxy equipment sends a plurality of acquired real-time operation pictures to the cloud server;
the cloud server inputs each real-time operation picture as input data into a second visual analysis model;
when the second visual analysis model outputs target operators and target operation positions which do not meet the operation requirements, determining the target operators and the target operation positions as formal analysis results, and sending the formal analysis results to the management terminal;
and when the output result of the second visual analysis model is empty, releasing the continuous acquisition instruction of the target ball machine to the target operator.
3. The engineering supervision method based on artificial intelligence data processing and visual analysis according to claim 1, wherein the step of determining a target ball machine for monitoring a target worker according to a target work position includes:
determining a ball set for enclosing the target operation position according to the target operation position and the installation position of each ball machine, wherein the ball set comprises a plurality of ball machines which jointly encircle the target operation position;
and controlling each ball machine of the ball machine set to adjust the shooting angle to face the target operator according to the target operation position so as to form the ball machine set for monitoring the target operator.
4. The engineering supervision method based on artificial intelligence data processing and visual analysis according to claim 3, wherein the step of continuously collecting real-time operation pictures of target operators by the control target ball machine in a set time period and sending the real-time operation pictures to the edge proxy equipment comprises the following steps:
forming a multi-angle monitoring image of a target operator through real-time operation images respectively acquired by each ball machine in the ball machine set;
the edge proxy equipment sends the multi-angle monitoring images at the same acquisition time point to different layers according to different visual angles;
scaling deformation is carried out on images with different visual angles in each image layer so as to fuse multi-angle monitoring images with the same acquisition time point to form a three-dimensional operation picture with the same acquisition time point.
5. The engineering supervision method based on artificial intelligence data processing and visual analysis according to claim 2, further comprising:
establishing an image library storing a plurality of target images, wherein the target images comprise multi-angle images in a standard working state;
simplifying the multi-angle image according to the standard operation state at the cloud server to obtain an abstract graph;
The abstract graph is sent to the first visual analysis model and the second visual analysis model to identify a standard job status from the input image by the first visual analysis model and the second visual analysis model.
6. The engineering supervision method based on artificial intelligence data processing and visual analysis according to claim 1, wherein the edge proxy device is provided with a processing module; the method further comprises the steps of:
acquiring the position of an operator detected by a human body infrared sensor and a shooting period corresponding to the position of each operator by a ball machine so as to form first monitoring data;
the processing module forms operation dynamic record data of operators according to the first monitoring data;
acquiring each target operator, a target operation position corresponding to each target operator and a corresponding shooting period to form second monitoring data;
and the processing module forms nonstandard operation data according to the second monitoring data, wherein the nonstandard operation data comprises personnel, operation positions and operation time periods which do not meet the operation requirements.
7. The engineering supervision method based on artificial intelligence data processing and visual analysis according to claim 6, further comprising:
After monitoring starts, the edge proxy equipment determines the start and end of the starting time of the operation picture according to the infrared data collected by the human body infrared sensor and the image collected by the dome camera;
the edge proxy equipment marks the time zone which falls into the operation picture starting time and the time zone which does not fall into the operation picture starting time respectively;
and storing the pictures acquired by the ball machine in the time section of the operation picture starting time and the pictures acquired by the ball machine in the time section of the operation picture starting time to different storage positions of the edge proxy equipment.
8. An engineering supervision system, characterized in that an engineering supervision method based on artificial intelligence data processing and visual analysis according to any one of claims 1 to 7 is applied, the engineering supervision system comprises a plurality of ball machines and a plurality of human body infrared sensors arranged in a monitoring range, the ball machines and the human body infrared sensors are respectively in communication connection with edge proxy equipment, and the edge proxy equipment is used for being in communication connection with a cloud server.
CN202310686725.XA 2023-06-12 2023-06-12 Engineering supervision method and system based on artificial intelligence data processing and visual analysis Active CN116437216B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310686725.XA CN116437216B (en) 2023-06-12 2023-06-12 Engineering supervision method and system based on artificial intelligence data processing and visual analysis

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310686725.XA CN116437216B (en) 2023-06-12 2023-06-12 Engineering supervision method and system based on artificial intelligence data processing and visual analysis

Publications (2)

Publication Number Publication Date
CN116437216A CN116437216A (en) 2023-07-14
CN116437216B true CN116437216B (en) 2023-09-08

Family

ID=87080054

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310686725.XA Active CN116437216B (en) 2023-06-12 2023-06-12 Engineering supervision method and system based on artificial intelligence data processing and visual analysis

Country Status (1)

Country Link
CN (1) CN116437216B (en)

Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012093912A (en) * 2010-10-26 2012-05-17 Secom Co Ltd Entering body detection device
DE202012103130U1 (en) * 2012-08-20 2012-09-18 Günther Ende Industrial camera system
WO2013052383A1 (en) * 2011-10-07 2013-04-11 Flir Systems, Inc. Smart surveillance camera systems and methods
JP2016015085A (en) * 2014-07-03 2016-01-28 真 高津 Personnel employment support system using video clip and device thereof
CN206863891U (en) * 2017-06-21 2018-01-09 山东壹云智能科技有限公司 A kind of seamless tracks of device of ball machine
CN206977597U (en) * 2017-07-28 2018-02-06 盛视科技股份有限公司 Video device for automatically capturing head portrait
WO2018076191A1 (en) * 2016-10-26 2018-05-03 深圳前海达闼云端智能科技有限公司 Smart patrol device, cloud control device, patrol method, control method, robot, controller, and non-transient computer readable storage medium
CN110809121A (en) * 2019-10-28 2020-02-18 广州顶新时代电子***科技有限公司 Security system based on infrared thermal imaging technology
WO2020057353A1 (en) * 2018-09-21 2020-03-26 深圳市九洲电器有限公司 Object tracking method based on high-speed ball, monitoring server, and video monitoring system
CN112954315A (en) * 2021-02-25 2021-06-11 深圳市中西视通科技有限公司 Image focusing measurement method and system for security camera
CN113225525A (en) * 2021-04-01 2021-08-06 北京戴纳实验科技有限公司 Indoor monitoring method and system
CN113506416A (en) * 2021-07-03 2021-10-15 山东省水利勘测设计院 Engineering abnormity early warning method and system based on intelligent visual analysis
WO2022161080A1 (en) * 2021-01-28 2022-08-04 华为技术有限公司 Photographic method and apparatus
CN114997582A (en) * 2022-04-29 2022-09-02 广东电网有限责任公司广州供电局 Early warning method and system for distribution network operation risk and computer equipment

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140118554A1 (en) * 2012-10-30 2014-05-01 Valentine A. Bucknor System of a Surveillance Camera For Identifying And Incapacitating Dangerous Intruders

Patent Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012093912A (en) * 2010-10-26 2012-05-17 Secom Co Ltd Entering body detection device
WO2013052383A1 (en) * 2011-10-07 2013-04-11 Flir Systems, Inc. Smart surveillance camera systems and methods
DE202012103130U1 (en) * 2012-08-20 2012-09-18 Günther Ende Industrial camera system
JP2016015085A (en) * 2014-07-03 2016-01-28 真 高津 Personnel employment support system using video clip and device thereof
WO2018076191A1 (en) * 2016-10-26 2018-05-03 深圳前海达闼云端智能科技有限公司 Smart patrol device, cloud control device, patrol method, control method, robot, controller, and non-transient computer readable storage medium
CN206863891U (en) * 2017-06-21 2018-01-09 山东壹云智能科技有限公司 A kind of seamless tracks of device of ball machine
CN206977597U (en) * 2017-07-28 2018-02-06 盛视科技股份有限公司 Video device for automatically capturing head portrait
WO2020057353A1 (en) * 2018-09-21 2020-03-26 深圳市九洲电器有限公司 Object tracking method based on high-speed ball, monitoring server, and video monitoring system
CN110809121A (en) * 2019-10-28 2020-02-18 广州顶新时代电子***科技有限公司 Security system based on infrared thermal imaging technology
WO2022161080A1 (en) * 2021-01-28 2022-08-04 华为技术有限公司 Photographic method and apparatus
CN112954315A (en) * 2021-02-25 2021-06-11 深圳市中西视通科技有限公司 Image focusing measurement method and system for security camera
CN113225525A (en) * 2021-04-01 2021-08-06 北京戴纳实验科技有限公司 Indoor monitoring method and system
CN113506416A (en) * 2021-07-03 2021-10-15 山东省水利勘测设计院 Engineering abnormity early warning method and system based on intelligent visual analysis
CN114997582A (en) * 2022-04-29 2022-09-02 广东电网有限责任公司广州供电局 Early warning method and system for distribution network operation risk and computer equipment

Also Published As

Publication number Publication date
CN116437216A (en) 2023-07-14

Similar Documents

Publication Publication Date Title
CN107911653B (en) Intelligent video monitoring module, system, method and storage medium for residence
CN112785798B (en) Behavior analysis method for constructors of power substation engineering construction project
CN112698676B (en) AI-based intelligent power distribution room operation method
CN108981106B (en) Hospital central air-conditioning safety three-dimensional monitoring cloud platform, system and method
CN207993197U (en) A kind of chemical industry smoke and fire intelligent video analytic system
CN213781155U (en) Cable pit calamity intelligent monitoring and processing system
CN113551775A (en) Equipment fault on-line monitoring and alarming method and system based on infrared thermal imaging
CN111932709A (en) Method for realizing violation safety supervision of inspection operation of gas station based on AI identification
CN111931652A (en) Dressing detection method and device and monitoring terminal
CN113206978A (en) Security intelligent monitoring early warning system and method for oil and gas pipeline station
CN112818758A (en) Monitoring method, system and device for electric power construction site and storage medium
CN112216059A (en) Fire control information management platform equipment based on thing networking
CN113506416A (en) Engineering abnormity early warning method and system based on intelligent visual analysis
CN116437216B (en) Engineering supervision method and system based on artificial intelligence data processing and visual analysis
CN115761591A (en) Multi-operation-site violation identification method
CN113378749B (en) Visual inspection system based on big data analysis
CN113596408A (en) Auxiliary monitoring system based on AI intelligent analysis
CN111160771A (en) Safety supervision system and control method thereof
CN110751125A (en) Wearing detection method and device
CN116311729A (en) AI-based power grid infrastructure site safety management system
CN112351108B (en) Real-time monitoring system suitable for ice rink
CN114697614A (en) Charging process safety monitoring system and method based on artificial intelligence
CN113392713A (en) Work clothes wearing identification method
CN113420739A (en) Intelligent emergency monitoring method and system based on neural network and readable storage medium
CN113538847A (en) Artificial intelligence monitoring device that can automatically regulated

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CP01 Change in the name or title of a patent holder
CP01 Change in the name or title of a patent holder

Address after: Room 605, Building 2, Huatie Jiayuan, No. 170, Youyi Road, Tianxin District, Changsha, 410000, Hunan Province

Patentee after: Hunan Bochuang High tech Industrial Co.,Ltd.

Address before: Room 605, Building 2, Huatie Jiayuan, No. 170, Youyi Road, Tianxin District, Changsha, 410000, Hunan Province

Patentee before: Hunan Boxin Chuangyuan Information Technology Co.,Ltd.