CN114937237A - Construction site safety monitoring method and system based on AI intelligent identification - Google Patents
Construction site safety monitoring method and system based on AI intelligent identification Download PDFInfo
- Publication number
- CN114937237A CN114937237A CN202210439452.4A CN202210439452A CN114937237A CN 114937237 A CN114937237 A CN 114937237A CN 202210439452 A CN202210439452 A CN 202210439452A CN 114937237 A CN114937237 A CN 114937237A
- Authority
- CN
- China
- Prior art keywords
- hidden danger
- information
- construction
- constructor
- constructors
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 241
- 238000000034 method Methods 0.000 title claims abstract description 62
- 238000012544 monitoring process Methods 0.000 title claims abstract description 49
- 230000002159 abnormal effect Effects 0.000 claims description 70
- 238000012216 screening Methods 0.000 claims description 38
- 238000001514 detection method Methods 0.000 claims description 26
- 210000000056 organ Anatomy 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 17
- 230000005856 abnormality Effects 0.000 claims description 16
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 231100001267 hazard identification Toxicity 0.000 claims 1
- 230000000977 initiatory effect Effects 0.000 claims 1
- 230000006399 behavior Effects 0.000 abstract description 13
- 230000000694 effects Effects 0.000 abstract description 3
- 230000000875 corresponding effect Effects 0.000 description 66
- 239000000779 smoke Substances 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000000391 smoking effect Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000009435 building construction Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- 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
- G08B21/18—Status alarms
- G08B21/24—Reminder alarms, e.g. anti-loss alarms
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Multimedia (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Computation (AREA)
- Emergency Management (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Educational Administration (AREA)
- Artificial Intelligence (AREA)
- Game Theory and Decision Science (AREA)
- Medical Informatics (AREA)
- Human Computer Interaction (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Social Psychology (AREA)
- Psychiatry (AREA)
- Primary Health Care (AREA)
- Alarm Systems (AREA)
- Emergency Alarm Devices (AREA)
Abstract
The application relates to a construction site safety monitoring method and system based on AI intelligent identification, which comprises the steps of obtaining construction image information of each construction position in a construction site, and identifying whether a hidden danger situation occurs in a construction image according to the construction image information; if yes, acquiring specific hidden danger type information and preset hidden danger level sequence list information in the hidden danger situation, and acquiring an actual hidden danger level corresponding to the specific hidden danger type from the hidden danger level sequence list; acquiring a preset notification mode sequence table, and acquiring an actual notification type corresponding to the actual hidden danger level from the notification mode sequence table; and acquiring the contact information of the constructors corresponding to the specific hidden danger types, and triggering a warning starting instruction according to the contact information of the constructors and the actual informing type. The method and the device have the effect of more efficiently informing constructors of existing potential safety hazard behaviors.
Description
Technical Field
The invention relates to the technical field of building construction safety management, in particular to a construction site safety monitoring method and system based on AI intelligent identification.
Background
At present, along with the continuous development of society, building construction site is constantly increasing, and wherein constructor is the important component part in the work progress to the security monitoring in the construction site for whether the action that detects constructor and whether there is the security threat, give constructor corresponding warning, provide the guarantee for constructor's construction security.
Present, constructor can be through shooing the constructor when getting into construction site, and the potential safety hazard action that the constructor exists is judged to the monitoring personnel, and the constructor is reminded to rethread loudspeaker for the monitoring personnel carry out the correspondence and correct, improve the security of constructor in construction site.
In view of the above-mentioned related technologies, the inventor thinks that there are defects that the number of constructors in a construction site is large and the efficiency of a manner of reminding the constructors is low.
Disclosure of Invention
In order to improve the effect that constructors are informed of existing potential safety hazard behaviors more efficiently, the application provides a construction site safety monitoring method and system based on AI intelligent identification.
The above object of the present invention is achieved by the following technical solutions:
a construction site safety monitoring method based on AI intelligent identification comprises the following steps:
acquiring construction image information of each construction position in a construction site, and identifying whether hidden danger situations occur in the construction images according to the construction image information;
if yes, acquiring specific hidden danger type information and preset hidden danger level sequence list information in the hidden danger situation, and acquiring an actual hidden danger level corresponding to the specific hidden danger type from the hidden danger level sequence list;
acquiring a preset notification mode sequence table, and acquiring an actual notification type corresponding to the actual hidden danger level from the notification mode sequence table;
and acquiring the contact information of the constructors corresponding to the specific hidden danger types, and triggering a warning starting instruction according to the contact information of the constructors and the actual informing type.
By adopting the technical scheme, the construction images are obtained from all construction positions in the construction site, whether hidden danger occurs at all the construction positions is identified from the construction images, and because monitoring equipment for monitoring all the construction positions is arranged in the construction site, the efficiency of obtaining all the construction images is high, and the obtaining cost is low; the hidden danger types of constructors in the construction process of a construction site are numerous, and the timeliness degree of each hidden danger type which needs to remind the constructors is different from the reminding mode adopted, so when the situation that the hidden danger exists in a construction image is identified, the actual hidden danger level corresponding to the specific hidden danger type is compared and analyzed from the hidden danger level sequence table, the severity of the specific hidden danger type can be obtained, and then the actual informing type is determined from the preset informing mode sequence table through the severity; then, according to the contact information of the constructors and the actual informing type, triggering an alarming starting instruction to remind the corresponding constructors, wherein the contact information of the constructors corresponds to the specific hidden danger type identified from the construction image; the potential safety hazard behavior can be improved and the constructors can be informed more efficiently.
The application may be further configured in a preferred example to: the hidden danger situations comprise self hidden dangers and scene hidden dangers, and the method for identifying whether the hidden danger situations occur in the construction image comprises the following steps:
identifying the position of a constructor from the construction image information, acquiring the position information of the key part of the constructor from the construction image information according to the position of the constructor, identifying hidden danger characteristics from the position information of the key part of the constructor, and taking the hidden danger characteristics as hidden dangers of the constructor;
identifying a hidden danger position from the construction image, calculating the distance between the position of a constructor and the hidden danger position, and judging whether the distance is smaller than a preset threshold value or not;
and if so, taking the behavior of the constructor approaching the position with the hidden danger as the scene hidden danger.
By adopting the technical scheme, the positions of the constructors are identified from the construction image, the positions of the key parts of the constructors are obtained from the positions of the constructors in the construction image, and the hidden danger characteristics are identified from the position information of the key parts of the constructors, are the hidden dangers of the constructors, so that targeted identification is achieved, and the accuracy of identifying the hidden dangers is improved; in addition, the position with hidden danger in the construction site environment is identified from the construction image, and then the distance between the position of the constructor and the position with hidden danger in the construction image can be used for determining whether the position of the constructor is influenced by the hidden danger in the construction environment, so that the behavior of the constructor approaching the position with hidden danger is taken as the scene hidden danger; the hidden danger identification situation identifies hidden dangers of construction personnel and environments where the construction personnel are located, so that the accuracy of potential safety hazard troubleshooting of the construction personnel in the construction process is greatly improved, and the safety of the construction personnel is further improved.
The present application may be further configured in a preferred example to: the hidden danger characteristics recognized from the position information of the key parts of the constructors are obtained by recognizing a judgment model, and the method for training the judgment model comprises the following steps:
training according to correct state data of the constructors in the historical data to obtain an abnormality detection model, wherein the abnormality detection model is used for judging whether the positions of all organs of the constructors are abnormal or not;
the method comprises the steps of obtaining construction site hidden danger type information, obtaining related constructor organ position information from each construction site hidden danger type, training according to the construction site hidden danger type information and the constructor organ position information to obtain a hidden danger model used for detecting specific hidden danger characteristics, and taking an abnormal detection model and a hidden danger model as judgment models.
By adopting the technical scheme, the abnormal detection model is obtained through multiple times of training according to the correct state data of the constructors in the historical data, so that whether the positions of the organs of the constructors are abnormal or not is preliminarily judged, the abnormal organs are screened out from the multiple organs of the constructors, and the complexity of subsequent detection on specific hidden danger types is reduced; and then, acquiring construction site hidden danger type information, acquiring related organ position information of the constructors aiming at each construction site hidden danger type, and training to obtain a hidden danger model for detecting each specific hidden danger characteristic, so that the hidden danger characteristic corresponding to the abnormal organ position of the constructors is obtained, and the detection accuracy is high.
The present application may be further configured in a preferred example to: the method for identifying and obtaining the hidden danger characteristics through the judgment model comprises the following steps:
inputting the construction image into an abnormality detection model for abnormality fuzzy judgment to obtain abnormal position information;
and pulling the hidden danger model to perform abnormal accurate judgment according to the abnormal position information to obtain the hidden danger characteristics.
By adopting the technical scheme, the shot construction image is input into an abnormity detection model for abnormity fuzzy judgment, namely, the position of an abnormal organ in a human body of a constructor is obtained firstly; and then, according to the abnormal organ position, pulling out the hidden danger model for accurate judgment, so that the hidden danger characteristic can be obtained, the efficiency of obtaining the hidden danger characteristic can be improved, and the accuracy of obtaining the hidden danger characteristic can be improved.
The application may be further configured in a preferred example to: the method for acquiring the contact information of the constructors corresponding to the specific hidden danger types comprises the following steps:
determining the constructors corresponding to the specific hidden danger types as abnormal constructors, and identifying the head characteristics of the abnormal constructors from the construction images;
intensively matching and inquiring the head characteristics in a preset constructor to obtain a matching and inquiring result;
and if the query is successful, taking the contact information corresponding to the queried constructor as the contact information of the abnormal constructor.
By adopting the technical scheme, because the most intuitive mode for judging the identity of one constructor is to identify through the head characteristics, the head characteristics corresponding to the constructor with specific hidden danger types are identified from the construction image information, then the head characteristics and the preset constructor are intensively matched and inquired to obtain a matching inquiry result, and if the matching inquiry result is successful, the contact information corresponding to the inquired constructor is used as the contact information of the abnormal constructor; the query mode is direct and efficient.
The present application may be further configured in a preferred example to: after the head features are subjected to matching query in a preset constructor set to obtain a matching query result, the method comprises the following steps:
if the query fails, acquiring preliminary screening contact information of all constructors in the construction image;
acquiring similarity information from the matching query result, sequencing the similarity information, and acquiring preliminary contact information of the constructors with the similarity within a preset range threshold according to a preset constructor set;
and comparing the preliminary screening contact information with the preliminary contact information, and taking the contact information which is simultaneously present in the preliminary screening contact information and the preliminary contact information as the contact information of the abnormal personnel.
By adopting the technical scheme, if the identities of abnormal workers cannot be inquired in a centralized matching manner from preset constructors, because most of the constructors exist in a construction site range shot by one construction image, the contact information of all the constructors is acquired from the construction image to obtain the primarily screened contact information, then the similarity information is acquired from the matching inquiry result, the similarities are sorted, and then the primary contact information of the corresponding constructors with the similarity within the preset range threshold is acquired; finally, comparing the preliminary screening contact information with the preliminary contact information, and taking the contact information which is simultaneously present in the preliminary screening contact information and the preliminary contact information as the contact information of the abnormal personnel; the judgment mode reduces the search range of preset construction personnel concentration, carries out targeted query and improves the efficiency.
The present application may be further configured in a preferred example to: the method for acquiring the preliminary screening contact information of all the people in the construction image comprises the following steps:
acquiring shooting range position information of the construction image, and acquiring mobile phone signal information in the shooting range position through a triangulation technology according to the shooting range position information;
and screening out the preliminary screening contact information according to the mobile phone signal information and a preset constructor set.
By adopting the technical scheme, the shooting range of each shooting device in the construction site is determined, so that the shooting range position information of the construction image on the construction site is obtained firstly; then, acquiring collected signal information in the shooting range position by three base stations arranged around a construction site through a triangulation positioning technology, acquiring information of registration of the mobile phone number by constructors corresponding to the mobile phone signals from a database through the acquired mobile phone signal information because most people carry mobile phones at present, and finally, intensively searching the mobile phone signal information in preset constructors to screen out preliminary screening contact information; the method for acquiring the preliminary screening contact information is direct, strong in pertinence and high in acquisition accuracy.
The second objective of the present invention is achieved by the following technical solutions:
a worksite safety monitoring system based on AI intelligent identification, the worksite safety monitoring system comprising:
the hidden danger identification module is used for acquiring construction image information of each construction position in a construction site and identifying whether a hidden danger situation occurs in a construction image according to the construction image information;
the hidden danger level determining module is used for acquiring specific hidden danger type information and preset hidden danger level sequence list information in a hidden danger situation and acquiring an actual hidden danger level corresponding to the specific hidden danger type from the hidden danger level sequence list if the hidden danger level determining module is used for acquiring the specific hidden danger type information and the preset hidden danger level sequence list information in the hidden danger situation;
the notification type determining module is used for acquiring a preset notification mode sequence table and acquiring an actual notification type corresponding to the actual hidden danger level from the notification mode sequence table;
and the warning starting triggering module is used for acquiring the contact information of the constructors corresponding to the specific hidden danger types and triggering a warning starting instruction according to the contact information of the constructors and the actual informing type.
By adopting the technical scheme, the construction image is obtained from each construction position in the construction site, whether hidden danger exists in each construction position is identified from the construction image, and because monitoring equipment for monitoring each construction position is arranged in the construction site, the efficiency of obtaining each construction image is high, and the obtaining cost is low; the hidden danger types of constructors in the construction process of a construction site are numerous, and the timeliness degree of each hidden danger type which needs to remind the constructors is different from the reminding mode adopted, so when the situation that the hidden danger exists in a construction image is identified, the actual hidden danger level corresponding to the specific hidden danger type is compared and analyzed from the hidden danger level sequence table, the severity of the specific hidden danger type can be obtained, and then the actual informing type is determined from the preset informing mode sequence table through the severity; then, according to the contact information of the constructors and the actual informing type, triggering an alarming starting instruction to remind the corresponding constructors, wherein the contact information of the constructors corresponds to the specific hidden danger type identified from the construction image; the potential safety hazard behavior can be improved and the constructors can be informed more efficiently.
In summary, the present application includes at least one of the following beneficial technical effects:
1. the construction image is obtained from each construction position in the construction site, whether hidden danger exists in each construction position is identified from the construction image, and monitoring equipment for monitoring each construction position is arranged in the construction site, so that the efficiency of obtaining each construction image is high, and the obtaining cost is low; the hidden danger types of constructors in the construction process of a construction site are numerous, and the timeliness degree of each hidden danger type which needs to remind the constructors is different from the reminding mode adopted, so when the situation that the hidden danger exists in a construction image is identified, the actual hidden danger level corresponding to the specific hidden danger type is compared and analyzed from the hidden danger level sequence table, the severity of the specific hidden danger type can be obtained, and then the actual informing type is determined from the preset informing mode sequence table through the severity; then, according to the contact information of the constructors and the actual informing type, triggering an alarming starting instruction to remind the corresponding constructors, wherein the contact information of the constructors corresponds to the specific hidden danger type identified from the construction image; the method can improve the condition that the existing potential safety hazard behaviors are more efficiently informed to constructors;
2. firstly, inputting a shot construction image into an abnormality detection model for abnormality fuzzy judgment, namely, firstly acquiring the position of an abnormal organ in a human body of a constructor; then, according to the position of the abnormal organ, pulling out the hidden danger model corresponding to the organ for accurate judgment, so that hidden danger characteristics can be obtained, the efficiency of obtaining the hidden danger characteristics can be improved, and the accuracy of obtaining the hidden danger characteristics can also be improved;
3. the method comprises the steps that because the shooting range of each shooting device in a construction site is determined, the shooting range position information of a construction image on the construction site is obtained firstly; then, acquiring collected signal information in the shooting range position by three base stations arranged around a construction site through a triangulation positioning technology, acquiring information of registration of the mobile phone number by constructors corresponding to the mobile phone signals from a database through the acquired mobile phone signal information because most people carry mobile phones at present, and finally, intensively searching the mobile phone signal information in preset constructors to screen out preliminary screening contact information; the method for acquiring the preliminary screening contact information is direct, strong in pertinence and high in acquisition accuracy.
Drawings
Fig. 1 is a flowchart of a method for monitoring safety of a construction site based on AI intelligent recognition according to an embodiment of the present application.
Fig. 2 is a flowchart illustrating an implementation of identifying whether a hidden danger situation occurs in a construction image in a construction site safety monitoring method based on AI intelligent identification in an embodiment of the present application.
FIG. 3 is a flowchart of a method for training a decision model in a method for monitoring safety of a construction site based on AI intelligent recognition according to an embodiment of the present application.
Fig. 4 is a flowchart illustrating an implementation of a method for determining hidden danger characteristics obtained by model recognition in a construction site safety monitoring method based on AI intelligent recognition in an embodiment of the present application.
Fig. 5 is a flowchart illustrating the implementation of step S40 in the method for monitoring safety of a construction site based on AI intelligent recognition according to an embodiment of the present application.
Fig. 6 is a flowchart illustrating an implementation of obtaining preliminary screening contact information of all constructors in a construction image in the construction site safety monitoring method based on AI intelligent recognition in an embodiment of the present application.
FIG. 7 is a functional block diagram of a worksite safety monitoring system based on AI intelligent identification in an embodiment of the present application.
Detailed Description
The present application is described in further detail below with reference to the attached drawings.
In one embodiment, as shown in fig. 1, the present application discloses a worksite safety monitoring method based on AI intelligent identification, which specifically includes the following steps:
s10: and acquiring construction image information of each construction position in the construction site, and identifying whether hidden danger situations occur in the construction images according to the construction image information.
In the embodiment, the hidden trouble situation is a situation which is not beneficial to the construction safety of constructors in a construction site.
Specifically, the construction condition of each construction position in the construction site is shot in real time through monitoring cameras distributed at each construction position of the construction site, so that construction image information is obtained; and then identifying whether hidden danger situations occur in the construction image, such as: the safety helmet is not worn by constructors, the constructors do not wear construction clothes, the constructors are close to the foundation pit, the constructors are close to the working area of the tower crane and the like.
S20: if yes, specific hidden danger type information and preset hidden danger level sequence list information in the hidden danger situation are obtained, and an actual hidden danger level corresponding to the specific hidden danger type is obtained from the hidden danger level sequence list.
In this embodiment, the specific hidden danger type refers to a hidden danger type corresponding to the hidden danger situation; the hidden danger level is used for identifying the danger degree of each specific hidden danger type in the current construction project; the preset hidden danger grade sequence table is a one-to-one corresponding sequence table formed by arranging each hidden danger type and the corresponding hidden danger grade.
Specifically, if the hidden danger situation is identified in the construction image, specific hidden danger type information in the hidden danger situation is acquired, for example: if the hidden danger situation is that the constructor does not wear the safety helmet, the specific hidden danger type is the safety helmet; if the hidden trouble situation is smoke extraction during construction of construction personnel, the specific hidden trouble type is smoke extraction; in the process of construction of a construction site by a constructor, because the importance degrees of various hidden dangers influencing the safety of the constructor are different, the time for solving the hidden dangers is also different, the preset hidden danger level sequence list information corresponding to the construction site is obtained, the specific hidden danger types are input into the preset hidden danger level sequence list for searching, and the corresponding actual hidden danger levels are obtained, for example: if the specific hidden danger category is the safety helmet, the actual hidden danger level is high; if the specific hidden danger type is smoking, the actual hidden danger grade is a medium grade; if the specific hidden danger type is the construction suit, the actual hidden danger level is low, so that the importance degree and the notification timeliness of the hidden danger situation can be known, and the efficiency of subsequently informing corresponding constructors is improved.
More specifically, the preset hidden danger level sequence table in this embodiment may adjust hidden danger levels corresponding to different specific hidden danger types according to different scenes of a construction site, for example: if the construction staff has a middle great influence on the smoking behavior of the construction site corresponding to the construction site, the hidden danger level corresponding to the specific hidden danger type of smoking can be adjusted to be high in the preset hidden danger level sequence list.
S30: and acquiring a preset informing mode sequence table, and acquiring an actual informing type corresponding to the actual hidden danger level from the informing mode sequence table.
In this embodiment, the preset notification manner sequence table refers to a notification manner corresponding to each hidden danger level.
Specifically, a preset notification mode sequence table is obtained, the hidden danger level is input into the preset notification mode sequence table for searching, and an actual notification type is obtained, for example: if the hidden danger level is high, the actual informing type is that a horn shouting mode is adopted to pertinently remind the constructor; if the hidden danger level is a middle level, the actual informing type is that the constructor and a manager of a corresponding department are contacted through a telephone; if the hidden danger level is low, the actual informing type is that the constructor and a corresponding department manager are reminded through short message informing; the horn sound-absorbing device not only can reduce the influence on the work of constructors due to the frequent occurrence of horn sound in construction sites, but also is more humanized.
More specifically, the preset notification mode sequence table in this embodiment may adjust the actual notification types corresponding to different hidden danger levels according to different construction site scenes, for example: if the construction site probably has the more serious condition than senior, this hidden danger level of multiplicable serious, its type of actually telling can be changed into the mode that all loudspeakers of adopting whole construction site shout simultaneously, reminds constructor to evacuate construction site etc..
S40: and acquiring the contact information of the constructors corresponding to the specific hidden danger types, and triggering a warning starting instruction according to the contact information of the constructors and the actual informing type.
In this embodiment, the contact information of the constructors refers to the contact information of the constructors corresponding to the specific hidden danger types generated in the construction image; the constructor contact information includes: personal header images, personal names, personal contact phones, part of administrator names and contact phones, etc.
Specifically, matching is carried out on constructors corresponding to the types of the generated specific hidden dangers in the construction image with constructors in the construction site, so that contact information of the constructors is obtained, and then a warning starting instruction is triggered according to the actual informing type, so that the corresponding constructors can correct the generated hidden dangers as soon as possible; for example: if the type is called for adopting the loudspeaker to call this constructor for the pertinence in the actual type of telling, then loudspeaker can call out: zhang III, please bring the safety helmet. ".
In an embodiment, the hidden danger situations include self hidden danger and scene hidden danger, and as shown in fig. 2, the method for identifying whether a hidden danger situation occurs in a construction image includes:
s11: identifying the position information of the constructors from the construction image information, acquiring the position information of key parts of the constructors from the construction image information according to the positions of the constructors, identifying hidden danger characteristics from the position information of the key parts of the constructors, and taking the hidden danger characteristics as hidden dangers of the constructors.
In this embodiment, the key part position information refers to the position of an organ that a constructor needs to perform safety protection; the hidden danger of the construction personnel refers to the hidden danger caused by insufficient protection of the construction personnel on the body.
Specifically, the position of a constructor is identified through an AI image identification function from a construction image, and then the position of a key part of the constructor is obtained from the position of the constructor corresponding to the construction image; and then, identifying potential hazard characteristics from the position information of the key parts of the constructors through a judgment model, and taking the potential hazard characteristics as self potential hazards, wherein the self potential hazards comprise: constructors do not wear safety helmets, do not wear construction clothes and the like; therefore, targeted identification is achieved, and the accuracy of hidden danger identification is improved.
S12: and identifying the position of the hidden danger from the construction image, calculating the distance between the position of the constructor and the position of the hidden danger, and judging whether the distance is smaller than a preset threshold value.
In this embodiment, the predetermined threshold value refers to a minimum safe distance that needs to be maintained between the position of the constructor and the position of the hidden trouble.
Specifically, during construction of a construction site, the position identification of potential safety hazards in the construction environment is updated in real time in a shooting range corresponding to a construction image; therefore, the hidden danger position is recognized according to the position identification in the construction image, the connection line identification is carried out between the position of the constructor and the hidden danger position in the horizontal direction, then the proportion of the connection line identification and the construction image in the horizontal direction is calculated, the actual length of the connection line identification is calculated through pixels according to the actual height and the proportion of the construction image in the horizontal direction, namely, the distance between the position of the constructor and the hidden danger position is calculated, for example: the actual length of the construction image in the horizontal direction is 20m, and the ratio of the connection line identifier to the construction image in the horizontal direction is 3: 10 (for example, the total number of pixels in the horizontal direction of the construction image is 900, and 270 pixels are arranged between the pixels at the highest position and the pixels at the lowest position of the crops in the longitudinal direction of the image, so that the ratio is 3: 10), and finally, the length of the connection line identifier is calculated to be 6m, namely, the distance between the position of the constructor and the position of the hidden danger is 6 m; and finally, comparing the distance with a preset threshold value, and judging whether the distance between the position of the constructor and the position of the hidden danger is less than the minimum safe distance.
S13: and if so, taking the behavior of the constructor approaching the position with the hidden danger as the scene hidden danger.
In this embodiment, the scene hidden danger refers to a hidden danger that may be generated due to a potential safety hazard caused by a construction environment at a construction site where a constructor is located.
Specifically, if the distance between the position of the constructor and the position of the hidden danger is judged to be less than the minimum safe distance, the behavior that the constructor is close to the position of the hidden danger is taken as the scene hidden danger, for example, the constructor is close to the behavior of the range position of the tower crane, so that the potential safety hazard situation existing in the construction process of the constructor due to the construction environment can be detected, and the detection range is improved.
In an embodiment, the hidden danger features identified from the position information of the key parts of the constructor are obtained by identifying a decision model, as shown in fig. 3, the method for training the decision model includes:
s111: and training according to the correct state data of the constructors in the historical data to obtain an abnormality detection model, wherein the abnormality detection model is used for judging whether the positions of all organs of the constructors are abnormal or not.
In this embodiment, the historical data refers to the whole body image data of different angles at which constructors stand in the construction process of different construction departments; the correct state data of the constructors refers to the data of the whole body images of different angles in which the constructors stand under the condition that the physical safety measures of the constructors are qualified aiming at the construction departments in the construction range corresponding to the construction images.
Specifically, under the condition that the body safety measures of the constructors are qualified, the characteristic information of the positions of the head, the face, the chest and the like of the constructors is obtained from the historical image data, namely, the states of correctly wearing helmets, having no smoke and correctly wearing construction suits by the constructors are corresponded, and then an abnormity detection model is obtained through training, so that the abnormity detection model can be used for judging whether the head of the constructors is abnormal or not, whether the mouths of the constructors are abnormal or not, and whether the chest of the constructors is abnormal or not, and can be used for preliminary judgment.
S112: the method comprises the steps of obtaining construction site hidden danger type information, obtaining related constructor organ position information from each construction site hidden danger type, training according to the construction site hidden danger type information and the constructor organ position information to obtain a hidden danger model used for detecting specific hidden danger characteristics, and taking an abnormality detection model and a hidden danger model as judgment models.
In this embodiment, the site potential trouble type information is a potential trouble type that may occur to a constructor in a construction site.
Specifically, the information of the site hidden danger category is acquired, for example: do not wear the safety helmet, smoke, do not wear construction clothes etc. then carry out the analysis to every building site hidden danger kind, acquire the constructor organ position that every building site hidden danger kind is correlated, for example: the hidden danger model can be used for accurately detecting that the hidden danger characteristic corresponding to the abnormal phenomenon exists at the top of the head is that the safety helmet is not worn, the hidden danger characteristic corresponding to the abnormal phenomenon exists at the mouth is that the safety helmet is not worn, and the hidden danger characteristic corresponding to the abnormal phenomenon exists at the front of the chest is that the construction clothes are not worn.
In an embodiment, as shown in fig. 4, the method for identifying hidden danger features by using a decision model includes:
s113: and inputting the construction image into an abnormality detection model for abnormality fuzzy judgment to obtain abnormal position information.
Specifically, after a construction image is acquired, the construction image is directly input into an abnormality detection model, and each organ at the position of a constructor in the construction image is subjected to fuzzy judgment to determine whether an abnormal phenomenon exists or not, so that abnormal position information is acquired; therefore, the range of obtaining hidden danger characteristics by identification is narrowed, and the identification efficiency is improved.
S114: and pulling the hidden danger model to perform abnormal accurate judgment according to the abnormal position information to obtain hidden danger characteristics.
Specifically, after the information of the abnormal position is obtained through fuzzy judgment, the hidden danger behavior corresponding to the abnormal position is judged through the hidden danger model, that is, the hidden danger characteristic is accurately judged.
In an embodiment, as shown in fig. 5, the method for acquiring the contact information of the constructor corresponding to the specific hidden danger category includes:
s41: and determining the constructors corresponding to the specific hidden danger types as abnormal constructors, and identifying the head characteristics of the abnormal constructors from the construction images.
Specifically, the constructor who concrete hidden danger kind corresponds will appear decides for the unusual personnel who have the hidden danger, because the identity of one person, can judge through the head characteristic more directly to through AI image recognition function, discern the head characteristic who reachs unusual personnel from the construction image, in order to be used for judging the identity of unusual personnel more directly.
S42: and intensively matching and inquiring the head characteristics in a preset constructor to obtain a matching and inquiring result.
In the present embodiment, the preset set of constructors means contact information of all constructors in the construction site.
Specifically, after the head features are identified, the head features are input into a preset constructor set, the head features and head images of all persons in the preset constructor set are compared and matched through similarity comparison, and then matching query results are obtained, wherein the matching query results comprise query success and query failure, namely the head features of abnormal persons are queried in the preset constructor set or the head features of the abnormal persons cannot be queried in the preset constructor set.
S43: and if the query is successful, taking the contact information corresponding to the queried constructor as the contact information of the abnormal constructor.
In this embodiment, when the head features of the abnormal staff are intensively inquired from the preset constructors, the inquired contact information corresponding to the constructors is used as the contact information of the abnormal staff, and then the potential hazards of the abnormal staff can be immediately reminded according to the actual informing type through the contact information, so that the corresponding constructors and the corresponding responsible staff can be informed more efficiently.
In one embodiment, as shown in fig. 5, after step S42, the method includes:
s44: and if the query fails, acquiring the preliminary screening contact information of all constructors in the construction image.
Specifically, when the head characteristics of the abnormal personnel cannot be inquired from the preset constructor set, the contact information of all constructors appearing in the construction image is obtained first, and the preliminary screening contact information is obtained, so that all constructors in the construction site are preliminarily screened, the determination range of the contact information of the abnormal personnel is reduced, and the efficiency is improved.
S45: and acquiring similarity information from the matching query result, sequencing the similarity information, and acquiring the preliminary contact information of the constructors with the similarity within a preset range threshold according to a preset constructor set.
Specifically, because the matching query is obtained by a similarity comparison mode, the similarity can be directly obtained from the matching query result, namely the similarity obtained after the head characteristics of the abnormal personnel and the personal head image of each constructor in a preset constructor set are obtained; and then, sequencing the similarity from high to low, comparing the similarity with a preset range threshold to obtain the similarity reaching the preset range threshold, and obtaining contact information of constructors corresponding to the similarity from a preset constructor set to obtain preliminary contact information, so that all constructors in a construction site can be preliminarily screened, the determination range of contact information of abnormal personnel is narrowed, and the efficiency is improved.
S46: and comparing the preliminary screening contact information with the preliminary contact information, and taking the contact information which is simultaneously present in the preliminary screening contact information and the preliminary contact information as the contact information of the abnormal personnel.
Specifically, after primary screening contact information and primary contact information are acquired aiming at the same construction image, the primary screening contact information and the primary contact information are compared, the contact information of the contact information appearing in the primary screening contact information and the primary contact information at the same time is compared, and the contact information is the same as the contact information above two contact information contents in the comparison process, all the contact information contents are not needed to be compared, the efficiency is improved, then the contact information appearing in the primary screening contact information and the primary contact information at the same time is used as the contact information of abnormal personnel, and therefore the accuracy of judging the contact information of the abnormal personnel is improved.
In an embodiment, as shown in fig. 6, the method for acquiring the preliminary screening contact information of all constructors in the construction image includes:
s441: and acquiring the shooting range position information of the construction image, and acquiring the mobile phone signal information in the shooting range position through a triangulation technology according to the shooting range position information.
Specifically, when each monitoring camera is installed, the range of each monitoring camera for shooting the construction site is calculated; at present, each constructor carries a mobile phone, and each mobile phone corresponds to a unique constructor, so when the head characteristics of abnormal personnel cannot be inquired from a preset constructor set, the shooting range position of a monitoring camera corresponding to a construction image is obtained; then accurately acquiring a mobile phone signal corresponding to each unique mobile phone identification in the shooting range position through three base stations arranged near a construction site by a triangulation technology, further directly acquiring the unique identification of a mobile phone SIM card to which the mobile phone signal belongs according to collected signal information to serve as the unique identification of the mobile phone, and then acquiring the mobile phone number of the mobile phone through the SIM card; therefore, the corresponding contact information of the constructors can be acquired from the preset constructors in a centralized manner through mobile phone signals only by adding the mobile phone numbers corresponding to the mobile phone SIM cards of the constructors in the preset constructors in a centralized manner, and the accuracy of the method for acquiring the preliminary screening contact information is high.
S442: and screening out the preliminary screening contact information according to the mobile phone signal information and a preset constructor set.
Specifically, after a mobile phone signal within a shooting range position in the construction image is obtained, mobile phone SIM card information to which the mobile phone signal belongs is obtained, then mobile phone numbers are obtained from a preset construction worker set according to the mobile phone SIM card, and then preliminary screening contact information can be screened out, and the method is high in accuracy and efficiency.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
In one embodiment, a worksite safety monitoring system based on AI intelligent identification is provided, and the worksite safety monitoring system based on AI intelligent identification corresponds to the worksite safety monitoring method based on AI intelligent identification in the above embodiments one to one. As shown in fig. 7, the AI intelligent recognition-based worksite safety monitoring system includes a hidden danger recognition module, a hidden danger level determination module, an informing type determination module, and an alarm start triggering module. The functional modules are explained in detail as follows:
the hidden danger identification module is used for acquiring construction image information of each construction position in a construction site and identifying whether a hidden danger situation occurs in a construction image according to the construction image information;
the hidden danger grade determining module is used for acquiring specific hidden danger type information and preset hidden danger grade sequence list information in the hidden danger situation and acquiring the actual hidden danger grade corresponding to the specific hidden danger type from the hidden danger grade sequence list if the hidden danger type is determined to be the specific hidden danger type;
the notification type determining module is used for acquiring a preset notification mode sequence table and acquiring an actual notification type corresponding to the actual hidden danger level from the notification mode sequence table;
and the warning starting triggering module is used for acquiring the contact information of the constructors corresponding to the specific hidden danger types and triggering a warning starting instruction according to the contact information of the constructors and the actual informing type.
Optionally, the hidden danger identifying module includes:
the self hidden danger determining submodule is used for identifying the position of a constructor from the construction image information, acquiring the position information of a key part of the constructor from the construction image information according to the position of the constructor, identifying hidden danger characteristics from the position information of the key part of the constructor and taking the hidden danger characteristics as self hidden dangers;
the hidden danger position identification submodule is used for identifying a hidden danger position from the construction image, calculating the distance between the position of a constructor and the hidden danger position and judging whether the distance is smaller than a preset threshold value or not;
and the scene hidden danger determining submodule is used for taking the behavior of the constructor approaching the hidden danger position as the scene hidden danger if the scene hidden danger determining submodule is used for determining the scene hidden danger.
Optionally, the building site safety monitoring system based on AI intelligent recognition further includes:
the anomaly detection model training module is used for training to obtain an anomaly detection model according to correct state data of the constructors in the historical data, and the anomaly detection model is used for judging whether each organ position of the constructors is abnormal or not;
the hidden danger model training module is used for acquiring construction site hidden danger type information, acquiring related constructor organ position information from each construction site hidden danger type, training according to the construction site hidden danger type information and the constructor organ position information to obtain a hidden danger model for detecting each specific hidden danger characteristic, and taking the abnormal detection model and the hidden danger model as a judgment model.
Optionally, the building site safety monitoring system based on AI intelligent recognition further includes:
the fuzzy judgment module is used for inputting the construction image into the abnormity detection model for abnormity fuzzy judgment to obtain abnormity position information;
and the accurate judgment module is used for pulling the hidden danger model to perform accurate abnormal judgment according to the abnormal position information to obtain hidden danger characteristics.
Optionally, the alert start triggering module includes:
the head characteristic identification submodule is used for determining the constructors corresponding to the specific hidden danger types as abnormal constructors and identifying the head characteristics of the abnormal constructors from the construction images;
the matching query submodule is used for intensively matching and querying the head characteristics in a preset constructor to obtain a matching query result;
and the matching query judgment sub-module is used for taking the contact information corresponding to the queried constructor as the contact information of the abnormal constructor if the query is successful.
Optionally, the alert start triggering module further includes:
the primary screening contact information acquisition sub-module is used for acquiring the primary screening contact information of all constructors in the construction image if the query fails;
the preliminary contact information acquisition sub-module is used for acquiring similarity information from the matching query result, sorting the similarity information, and acquiring preliminary contact information of constructors with the similarity within a preset range threshold value according to a preset constructor set;
and the information comparison submodule is used for comparing the preliminary screening contact information with the preliminary contact information and taking the contact information which is simultaneously present in the preliminary screening contact information and the preliminary contact information as the contact information of the abnormal personnel.
Optionally, the preliminary screening contact information obtaining sub-module includes:
the mobile phone signal acquisition unit is used for acquiring the shooting range position information of the construction image and acquiring mobile phone signal information in the shooting range position through a triangulation technology according to the shooting range position information;
and the contact information screening unit is used for screening out the primarily screened contact information according to the mobile phone signal information and a preset constructor set.
For specific limitations of the AI-based smart identification-based worksite safety monitoring system, reference may be made to the above limitations of the AI-based smart identification-based worksite safety monitoring method, which are not described in detail herein. The modules in the building site safety monitoring system based on AI intelligent identification can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (10)
1. A construction site safety monitoring method based on AI intelligent recognition is characterized by comprising the following steps:
acquiring construction image information of each construction position in a construction site, and identifying whether hidden danger situations occur in the construction images according to the construction image information;
if yes, acquiring specific hidden danger type information and preset hidden danger level sequence list information in the hidden danger situation, and acquiring an actual hidden danger level corresponding to the specific hidden danger type from the hidden danger level sequence list;
acquiring a preset notification mode sequence table, and acquiring an actual notification type corresponding to the actual hidden danger level from the notification mode sequence table;
and acquiring the contact information of the constructors corresponding to the specific hidden danger types, and triggering a warning starting instruction according to the contact information of the constructors and the actual informing type.
2. The AI intelligent recognition based site safety monitoring method of claim 1 wherein the potential risk situations include self potential risk and scene potential risk, and the method of identifying whether a potential risk situation occurs in the construction image comprises:
identifying the position of a constructor from the construction image information, acquiring the position information of a key part of the constructor from the construction image information according to the position of the constructor, identifying hidden danger characteristics from the position information of the key part of the constructor, and taking the hidden danger characteristics as hidden dangers of the constructor;
identifying hidden danger positions from the construction image, calculating the distance between the positions of constructors and the hidden danger positions, and judging whether the distance is smaller than a preset threshold value or not;
and if so, taking the behavior of the constructor approaching the position with the hidden danger as the scene hidden danger.
3. The AI-based intelligent site safety monitoring method of claim 2, wherein the identification of the hidden danger characteristics from the position information of the key parts of the constructor is carried out by identification of a decision model, and the method for training the decision model comprises the following steps:
training according to correct state data of the constructors in the historical data to obtain an abnormality detection model, wherein the abnormality detection model is used for judging whether the positions of all organs of the constructors are abnormal or not;
the method comprises the steps of obtaining construction site hidden danger type information, obtaining related constructor organ position information from each construction site hidden danger type, training according to the construction site hidden danger type information and the constructor organ position information to obtain a hidden danger model used for detecting specific hidden danger characteristics, and taking an abnormal detection model and a hidden danger model as judgment models.
4. The AI-intelligent-recognition-based worksite safety monitoring method of claim 3, wherein the method for identifying the potential hazard characteristics by the decision model comprises:
inputting the construction image into an abnormality detection model for abnormality fuzzy judgment to obtain abnormal position information;
and pulling the hidden danger model to perform abnormal accurate judgment according to the abnormal position information to obtain the hidden danger characteristics.
5. The AI-intelligent-recognition-based worksite safety monitoring method of claim 1, wherein the method of obtaining the contact information of the constructors corresponding to the specific hidden danger category comprises:
determining the constructors corresponding to the specific hidden danger types as abnormal constructors, and identifying the head characteristics of the abnormal constructors from the construction images;
intensively matching and inquiring the head characteristics in a preset constructor to obtain a matching and inquiring result;
and if the query is successful, taking the contact information corresponding to the queried constructor as the contact information of the abnormal constructor.
6. The AI intelligent recognition based worksite safety monitoring method of claim 5, wherein after matching the head characteristics to a query in a preset set of constructors to obtain a matching query result, the method comprises:
if the query fails, acquiring the preliminary screening contact information of all constructors in the construction image;
acquiring similarity information from the matching query result, sequencing the similarity information, and acquiring preliminary contact information of the constructors with the similarity within a preset range threshold according to a preset constructor set;
and comparing the preliminary screening contact information with the preliminary contact information, and taking the contact information which is simultaneously present in the preliminary screening contact information and the preliminary contact information as the contact information of the abnormal personnel.
7. The AI-intelligent-recognition-based worksite safety monitoring method of claim 6, wherein the method of obtaining prescreening contact information for all constructors in the construction image comprises:
acquiring shooting range position information of the construction image, and acquiring mobile phone signal information in the shooting range position through a triangulation technology according to the shooting range position information;
and screening out the preliminary screening contact information according to the mobile phone signal information and a preset constructor set.
8. A worksite safety monitoring system based on AI intelligent recognition, the worksite safety monitoring system comprising:
the hidden danger identification module is used for acquiring construction image information of each construction position in a construction site and identifying whether a hidden danger situation occurs in a construction image according to the construction image information;
the hidden danger level determining module is used for acquiring specific hidden danger type information and preset hidden danger level sequence list information in a hidden danger situation and acquiring an actual hidden danger level corresponding to the specific hidden danger type from the hidden danger level sequence list if the hidden danger level determining module is used for acquiring the specific hidden danger type information and the preset hidden danger level sequence list information in the hidden danger situation;
the notification type determining module is used for acquiring a preset notification mode sequence table and acquiring an actual notification type corresponding to the actual hidden danger level from the notification mode sequence table;
and the warning starting triggering module is used for acquiring the contact information of the constructors corresponding to the specific hidden danger types and triggering a warning starting instruction according to the contact information of the constructors and the actual informing type.
9. The AI intelligent recognition-based worksite safety monitoring system of claim 8, wherein the potential hazard identification module comprises:
the self hidden danger determining submodule is used for identifying the position of a constructor from the construction image information, acquiring the position information of a key part of the constructor from the construction image information according to the position of the constructor, identifying hidden danger characteristics from the position information of the key part of the constructor, and taking the hidden danger characteristics as self hidden dangers;
the hidden danger position identification submodule is used for identifying a hidden danger position from the construction image, calculating the distance between the position of a constructor and the hidden danger position and judging whether the distance is smaller than a preset threshold value or not;
and the scene hidden danger determining submodule is used for taking the behavior of the constructor approaching the hidden danger position as the scene hidden danger if the scene hidden danger determining submodule is used for determining the scene hidden danger.
10. The AI-intelligent-identification-based worksite safety monitoring system of claim 8, wherein the alert initiation triggering module comprises:
the head feature identification submodule is used for determining constructors corresponding to the specific hidden danger types as abnormal personnel and identifying the head features of the abnormal personnel from the construction images;
the matching query submodule is used for intensively matching and querying the head characteristics in a preset constructor to obtain a matching query result;
and the matching query judgment sub-module is used for taking the contact information corresponding to the queried constructor as the contact information of the abnormal constructor if the query is successful.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210439452.4A CN114937237A (en) | 2022-04-25 | 2022-04-25 | Construction site safety monitoring method and system based on AI intelligent identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210439452.4A CN114937237A (en) | 2022-04-25 | 2022-04-25 | Construction site safety monitoring method and system based on AI intelligent identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114937237A true CN114937237A (en) | 2022-08-23 |
Family
ID=82861637
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210439452.4A Pending CN114937237A (en) | 2022-04-25 | 2022-04-25 | Construction site safety monitoring method and system based on AI intelligent identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114937237A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7317264B1 (en) | 2022-10-18 | 2023-07-28 | 三菱電機株式会社 | Risk value calculation device and risk value calculation system |
| CN116740819A (en) * | 2023-08-14 | 2023-09-12 | 中通信息服务有限公司 | Method and device for identifying construction abnormal behavior in construction site based on AI algorithm |
| CN116895142A (en) * | 2023-09-11 | 2023-10-17 | 关天建设工程有限公司 | Global monitoring method and system for building construction |
-
2022
- 2022-04-25 CN CN202210439452.4A patent/CN114937237A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7317264B1 (en) | 2022-10-18 | 2023-07-28 | 三菱電機株式会社 | Risk value calculation device and risk value calculation system |
| CN116740819A (en) * | 2023-08-14 | 2023-09-12 | 中通信息服务有限公司 | Method and device for identifying construction abnormal behavior in construction site based on AI algorithm |
| CN116740819B (en) * | 2023-08-14 | 2023-12-19 | 中通信息服务有限公司 | Method and device for identifying construction abnormal behavior in construction site based on AI algorithm |
| CN116895142A (en) * | 2023-09-11 | 2023-10-17 | 关天建设工程有限公司 | Global monitoring method and system for building construction |
| CN116895142B (en) * | 2023-09-11 | 2023-11-24 | 关天建设工程有限公司 | Global monitoring method and system for building construction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114937237A (en) | Construction site safety monitoring method and system based on AI intelligent identification | |
| JP4924607B2 (en) | Suspicious behavior detection apparatus and method, program, and recording medium | |
| CN110245630B (en) | Monitoring data processing method and device and readable storage medium | |
| CN109996182B (en) | Positioning method, device and system based on combination of UWB positioning and monitoring | |
| CN111428617A (en) | Video image-based distribution network violation maintenance behavior identification method and system | |
| CN113011833A (en) | Safety management method and device for construction site, computer equipment and storage medium | |
| CN113160509B (en) | Risk sensing method and system suitable for communities and construction sites | |
| CN111783530A (en) | Safety system and method for monitoring and identifying behaviors in restricted area | |
| CN112349015B (en) | Shelter management and control system and method based on Internet of things | |
| CN110717357B (en) | Early warning method and device, electronic equipment and storage medium | |
| CN113052107A (en) | Method for detecting wearing condition of safety helmet, computer equipment and storage medium | |
| KR20160074208A (en) | System and method for providing safety service using beacon signals | |
| CN113343854A (en) | Fire operation flow compliance detection method based on video monitoring | |
| CN111460985A (en) | On-site worker track statistical method and system based on cross-camera human body matching | |
| CN113223046A (en) | Method and system for identifying prisoner behaviors | |
| RU2315352C2 (en) | Method and system for automatically finding three-dimensional images | |
| CN112288975A (en) | Event early warning method and device | |
| CN109345787A (en) | A kind of anti-outer damage monitoring and alarming system of the transmission line of electricity based on intelligent image identification technology | |
| KR102142315B1 (en) | ATM security system based on image analyses and the method thereof | |
| CN114092875A (en) | Operation site safety supervision method and device based on machine learning | |
| CN112330742A (en) | Method and device for recording activity routes of key personnel in public area | |
| CN109460714B (en) | Method, system and device for identifying object | |
| CN111126100A (en) | Alarm method, alarm device, electronic equipment and storage medium | |
| CN113052049B (en) | Off-duty detection method and device based on artificial intelligent tool identification | |
| CN113297976B (en) | Base station intrusion detection method and system based on deep learning |
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 |