CN114612866A - Intelligent identification method, device and equipment for safety in building site - Google Patents
Intelligent identification method, device and equipment for safety in building site Download PDFInfo
- Publication number
- CN114612866A CN114612866A CN202210511276.0A CN202210511276A CN114612866A CN 114612866 A CN114612866 A CN 114612866A CN 202210511276 A CN202210511276 A CN 202210511276A CN 114612866 A CN114612866 A CN 114612866A
- Authority
- CN
- China
- Prior art keywords
- area
- danger
- image data
- model
- object classification
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Emergency Alarm Devices (AREA)
Abstract
The invention belongs to the field of building safety identification, and relates to a method, a device and equipment for intelligently identifying safety in a building site, which comprises the following steps: identifying the area where the second image data is located, setting a risk coefficient for each area, marking the area as a dangerous area when the risk coefficient is greater than a preset risk coefficient threshold value, and obtaining the position of the dangerous area; calculating an actual danger range based on the position of the danger area; obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value; according to the invention, the effective working range of the building equipment is obtained through the big data network, the actual danger range is calculated, the safety assessment precision in the building site is higher, the closer the constructor is to the effective working range of the building equipment, the higher the danger level warning is sent out, and the safety accidents caused by the construction site noise and the synchronous influence of the visual field blind area in the construction process can be effectively avoided.
Description
Technical Field
The invention relates to the field of building safety identification, in particular to a method, a device and equipment for intelligently identifying safety in a building site.
Background
For the research on the construction safety risk accidents, scholars at home and abroad carry out long and numerous researches, most of the scholars cannot directly and effectively guide the safety management of a construction site, and most of the research fields are the mechanism analysis of the cause after the accidents occur and the preventive measures of the accidents occur; relatively few researches on safety risk accident early warning are carried out, the early researches are American experts in the 60 th 20 th century, the early researches are put forward on the basis of combining crisis and risk management two theories, more early researches are applied to national safety early warning, and the early researches are gradually developed into enterprise management in long-term development; although relevant experts and scholars in China have certain research on early warning of construction safety risks.
How to eliminate the construction safety risk accident in the middle of sprouting, can direct the safety control of job site of efficient guidance, reduce the emergence of incident and guarantee labourer's life safety, reduce the benefit level that construction safety lost improved the enterprise is worth the problem of deep attention and research. Due to the reasons of blind areas of vision and the like, the probability value of safety accidents when constructors work under the dangerous factors is very high.
Disclosure of Invention
The invention overcomes the defects of the prior art and provides a method, a device and equipment for intelligently identifying the safety in a building site.
In order to achieve the purpose, the technical scheme of the invention is as follows:
the invention provides an intelligent identification method for safety in a building site, which is characterized by comprising the following steps:
acquiring first image data in a building site;
constructing an object classification model, feeding back the first image data to the object classification model in real time, and performing object classification on the first image data in the object classification model to obtain second image data;
identifying the area where the second image data is located, acquiring the risk coefficient of each area, marking the area as a dangerous area when the risk coefficient is larger than a preset risk coefficient threshold value, and acquiring the position of the dangerous area;
calculating an actual danger range based on the position of the danger area;
obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and determining a danger grade based on the displacement, wherein the danger grade is divided into a low danger grade, a medium danger grade and a high danger grade, and if the danger grade is higher than the medium danger grade, immediately sending out an early warning signal.
Further, in a preferred embodiment of the present invention, the calculating the actual danger range based on the position of the dangerous area specifically includes the following steps:
acquiring the effective working area range of the building equipment through a big data network;
acquiring a position coordinate value of the dangerous area;
obtaining a plurality of limit coordinate values according to the position coordinate values of the dangerous area and the effective working area range of the construction equipment;
and taking an area surrounded by the limit coordinate values as an actual danger range.
Further, in a preferred embodiment of the present invention, the calculating a displacement amount according to the actual risk range and the first coordinate value specifically includes:
and calculating a plurality of displacement amounts according to the actual danger range and the first coordinate value, screening out the minimum displacement amount from the plurality of displacement amounts, and taking the minimum displacement amount as the finally output displacement amount.
Further, in a preferred embodiment of the present invention, the step of classifying the first image data to obtain the second image data further includes the following steps:
constructing a safety helmet wearing model, and obtaining a safety helmet identification model and a human body identification model from the object classification model;
guiding the safety helmet identification model and the human body identification model into a safety helmet wearing model for training to obtain a trained safety helmet wearing model;
importing the second image data into the trained safety helmet wearing model to obtain a similar probability value;
judging whether the similar probability value is smaller than a preset probability value or not;
if the value is less than the preset value, a warning that the safety helmet is not worn is sent.
Further, in a preferred embodiment of the present invention, the importing the second image data into the trained safety helmet wearing model to obtain a probability value includes the following steps:
judging whether a human body recognition model exists in the second image data;
if the human body recognition model exists, the human body recognition model is imported into the trained safety helmet wearing model, and a similar probability value is calculated.
Further, in a preferred embodiment of the present invention, identifying a region where the second image data is located, obtaining a risk coefficient of each region, marking the region as a dangerous region when the risk coefficient is greater than a preset risk coefficient threshold, and obtaining a position of the dangerous region specifically includes the following steps:
setting a danger coefficient according to the type of the second image data, if the danger coefficient is equal to a preset danger coefficient, marking an area where the second image data is located, and setting the area as a behavior monitoring area;
acquiring the behavior state of the behavior monitoring area within a preset time, wherein the behavior state comprises a working state and a non-working state;
and if the behavior state is the working state, taking the behavior monitoring area as a dangerous area, and acquiring the position of the dangerous area.
The invention provides a device for intelligently identifying safety in a building site, which comprises a memory and a processor, wherein the memory comprises a program of an intelligent identification method for safety in the building site, and when the program of the intelligent identification method for safety in the building site is executed by the processor, the following steps are realized:
acquiring first image data in a building site;
constructing an object classification model, feeding back the first image data to the object classification model in real time, and performing object classification on the first image data in the object classification model to obtain second image data;
identifying the area where the second image data is located, acquiring the risk coefficient of each area, marking the area as a dangerous area when the risk coefficient is larger than a preset risk coefficient threshold value, and acquiring the position of the dangerous area;
calculating an actual danger range based on the position of the danger area;
obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and determining a danger grade based on the displacement, wherein the danger grade is divided into a low danger grade, a medium danger grade and a high danger grade, and if the danger grade is higher than the medium danger grade, immediately sending out an early warning signal.
Further, in a preferred embodiment of the present invention, the calculating the actual danger range based on the position of the dangerous area specifically includes the following steps:
acquiring the effective working area range of the building equipment through a big data network;
acquiring a position coordinate value of the dangerous area;
obtaining a plurality of limit coordinate values according to the position coordinate values of the dangerous area and the effective working area range of the construction equipment;
and taking an area surrounded by the limit coordinate values as an actual dangerous range.
Further, in a preferred embodiment of the present invention, the step of classifying the first image data to obtain the second image data further includes the following steps:
constructing a safety helmet wearing model, and obtaining a safety helmet identification model and a human body identification model from the object classification model;
guiding the safety helmet identification model and the human body identification model into a safety helmet wearing model for training to obtain a trained safety helmet wearing model;
importing the second image data into the trained safety helmet wearing model to obtain a similar probability value;
judging whether the similar probability value is smaller than a preset probability value or not;
if the value is less than the preset value, a warning that the safety helmet is not worn is sent.
A third aspect of the present invention provides an intelligent identification device for security in a building site, including:
the acquisition module acquires first image data in a building site;
the classification module is used for constructing an object classification model, feeding back the first image data to the object classification model in real time, and performing object classification on the first image data in the object classification model to obtain second image data;
the identification module is used for identifying the area where the second image data is located, acquiring the danger coefficient of each area, marking the area as a dangerous area when the danger coefficient is larger than a preset danger coefficient threshold value, and acquiring the position of the dangerous area;
the first calculation module is used for calculating an actual danger range based on the position of the danger area;
the second calculation module is used for obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and the early warning module is used for determining a danger level based on the displacement, the danger level is divided into a low danger level, a medium danger level and a high danger level, and if the danger level is higher than the medium danger level, an early warning signal is immediately sent out.
The invention solves the defects in the background technology and can achieve the following technical effects: according to the invention, the effective working range of the building equipment is obtained in the big data network, so that the actual danger range in the monitoring area is calculated, the safety assessment precision in the building site is higher, the closer the constructor is to the effective working range of the building equipment, the higher the danger level warning is sent, the safety accidents caused by noise of the construction site and synchronous influence of the visual field blind area in the construction process can be effectively avoided, and the probability value of the safety accidents in the dust-haze weather or the fog weather can be reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings of the embodiments can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a specific method of intelligent identification of security in a building site;
FIG. 2 is a flow chart of a particular method of calculating an actual hazard range;
FIG. 3 illustrates a flow chart of a method of determining whether a constructor is wearing a helmet;
FIG. 4 shows a flow chart of a method of determining a hazardous area;
FIG. 5 is a schematic diagram of a safety helmet of the intelligent safety identification method in a building site;
FIG. 6 illustrates a system block diagram of a security intelligent recognition device in a building site;
fig. 7 shows a block diagram of a security intelligent recognition device in a building site.
In the figure:
1. the device comprises a warning unit, a 2 communication unit and a 3 positioning unit.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
FIG. 1 is a flow chart of a specific method of intelligent identification of security in a building site;
the invention provides an intelligent identification method for safety in a building site, which is characterized by comprising the following steps:
s102, collecting first image data in a building site;
s104, constructing an object classification model, feeding back the first image data to the object classification model in real time, and performing object classification on the first image data in the object classification model to obtain second image data;
s106, identifying the area where the second image data is located, acquiring the risk coefficient of each area, marking the area as a dangerous area when the risk coefficient is larger than a preset risk coefficient threshold value, and acquiring the position of the dangerous area;
s108, calculating an actual danger range based on the position of the danger area;
s110, obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and S112, determining danger levels based on the displacement, wherein the danger levels are divided into a low danger level, a medium danger level and a high danger level, and if the danger levels are higher than the medium danger level, immediately sending out an early warning signal.
It should be noted that the camera is an infrared camera, and an object classification model is used to classify images acquired in a monitored area, wherein various types of images are trained by using a neural network to obtain a trained object classification model, the object classification model is classified by using a shape feature classification method of the images, and after sharpening of the images, details of human body contours and building equipment contours in the images are enhanced, for example, sharpening is performed by using an ABF algorithm and a GLAS algorithm, so that the human body contours and the building equipment contours are obtained, and further the shape contours in the images are classified, and the trained object classification model satisfies the following requirements:
wherein M is an object model value; l is a geometric characteristic coefficient and takes the value of 1; n is the number of contour points; i is the ith contour point; k is a core expression coefficient, and the highest likelihood of the color of the pigment in the contour is ensured;
pixel values within the contour;
is the average pixel value within the contour; b is the pixel brightness value of the contour point; c is a corrected brightness coefficient;
is the sum of the pixel values within the contour; z is the smoothness of the profile.
The main contour in the image is trained by utilizing the classification model, so that the image collected in real time is guided into the model to obtain a model corresponding to each image, and the real-time model and the training model are subjected to similar estimation, wherein the similar estimation meets the following requirements:
wherein S is a Gaussian variance value; p is a similar probability value; m is the value of the object model after training is finished,
is a real-time object model value;
it should be noted that the higher the probability that the calculated value of P is closer to 1 indicates that the target object is a target object type, and the closer the calculated value of P is to 0 indicates that the target object is not a target object type in the training model, where the gaussian variance value is obtained from the measurement instrument when the image is subjected to the noise reduction processing.
It should be noted that, in the implementation process, the monitoring site may be divided into a plurality of monitoring areas, and each monitoring area is provided with a plurality of cameras. On the other hand, the position of the target object image can be acquired in various ways, for example, by means of a camera provided with a positioning instrument in a monitoring area (at this time, the monitoring position of the camera is the position of the target object image), by means of a mode of setting the positioning instrument on a helmet of a constructor, setting the positioning instrument on a construction instrument, and the like, when the construction instrument (such as an excavator, a truck, a cargo platform, a forklift, and the like) appears in the monitoring area, a safety coefficient (assigned as 1) is added to the monitoring area at this time, which indicates that the danger probability value generated in the area is high; when no construction equipment is present in the monitored area, a safety factor (assigned as 0) is added to the area, which indicates that the probability value of danger occurrence in the area is low. The monitoring area where the construction equipment is located is updated in time no matter the construction equipment moves to any position of the construction site. Be close to the architectural instrument more as constructor, the regional dangerous degree in suggestion the place ahead is high, reminds constructor in advance through the speaker that sets up at safety helmet, and this incident that not only can effectively avoid bringing because the synchronous influence of the noisy and the field of vision blind area of building site can reduce the probability value that takes place the incident in grey haze weather or fog weather moreover.
FIG. 2 is a flow chart of a particular method of calculating an actual hazard range;
further, in a preferred embodiment of the present invention, the calculating the actual danger range based on the position of the dangerous area specifically includes the following steps:
s202, acquiring the effective working area range of the building equipment through a big data network;
s204, acquiring the position coordinate value of the dangerous area;
s206, obtaining a plurality of limit coordinate values according to the position coordinate values of the dangerous area and the effective working area range of the construction equipment;
and S208, taking the area surrounded by the limit coordinate values as an actual danger range.
It should be noted that, when the object classification model obtains a construction equipment, an effective working range of the construction equipment, such as an effective area of an excavator during fixed-point working, an effective working area of a vibration pile driver, an effective working area of a concrete delivery pump, and the like, is obtained from the big data network, each type of construction machine has a certain effective working range during fixed-point working, and the working range of a limit position is taken as a dangerous range, such as the effective working range of the excavator: a cylinder surrounded by the maximum arm length of the excavator, a plane rectangular coordinate system is established with a driving shaft of a driving position (such as the maximum arm length of a working arm driving the excavator), and the rotating shaft is 0 datum point, and the limit coordinate value can be expressed as (Rsin theta, Rcos theta), wherein R is the maximum arm length of the working arm of the excavator, and theta is a circumferential angle formed between the rotating shaft and the rotating arm. Determining a position coordinate value of the dangerous area according to the position of the image shot by the monitoring camera, wherein the coordinate value can be expressed as
Where k is the kth limit coordinate value, the limit coordinate value of the actual danger range of the construction equipment can be expressed as
And the area enclosed by each limit coordinate value is the actual dangerous area. The remaining limit coordinate values can be obtained in this similar manner. Since construction equipment is subject to change over time, the actual risk range also follows.
Further, in a preferred embodiment of the present invention, the calculating a displacement amount according to the actual risk range and the first coordinate value specifically includes:
and calculating a plurality of displacement amounts according to the actual danger range and the first coordinate value, screening out the minimum displacement amount from the plurality of displacement amounts, and taking the minimum displacement amount as the finally output displacement amount.
It should be noted that the actual displacement can be obtained by calculating the limit coordinate value of the working limit of the construction equipment and the coordinate value of the current monitoring area of the constructor; because the current actual working range of building apparatus is a circular work area or fan-shaped work area, thereby can obtain a plurality of displacement according to the calculation between coordinate and the coordinate, select one of them minimum displacement as the displacement of final output, it is when this displacement is less than the predetermined displacement, accessible alarm device (install the speaker like the safety helmet on the constructor, the speaker in the building place) suggestion constructor the place ahead has danger, constructor can in time obtain the suggestion, this can avoid leading to the incident emergence of accident because the blind area of field of vision, reduce the emergence of accident.
FIG. 3 illustrates a flow chart of a method of determining whether a constructor is wearing a helmet;
further, in a preferred embodiment of the present invention, the step of classifying the first image data to obtain the second image data further includes the following steps:
s302, constructing a safety helmet wearing model, and obtaining a safety helmet identification model and a human body identification model from the object classification model;
s304, leading the safety helmet identification model and the human body identification model into a safety helmet wearing model for training to obtain a trained safety helmet wearing model;
s306, importing the second image data into the trained safety helmet wearing model to obtain a similar probability value;
s308, judging whether the similar probability value is smaller than a preset probability value or not;
and S310, if the current value is less than the preset value, giving out a warning that the safety helmet is not worn.
It should be noted that the safety helmet model can be constructed according to the modeling mode of the object classification model, that is, the safety helmet model and the human body recognition model are integrated, and the head condition of the constructor can be collected through the camera, so that whether the constructor wears the helmet or not can be judged. On the other hand, whether the wearing is standard wearing is obtained according to the similar estimation calculation mode, when the probability value is close to 1, the wearing standard of the safety helmet of the construction personnel is indicated, when the probability value is close to 0, the head of the construction personnel is not worn or the wearing of the safety helmet is not standard, and at the moment, the safety helmet can be provided with a loudspeaker to inform the construction personnel to wear the safety helmet when entering the construction area or prompt the construction personnel to wear the safety helmet regularly. When prompting constructors to wear the safety helmet regularly, if a pressure sensor is arranged in a fastening belt in the safety helmet, when the safety helmet is not fastened, a loudspeaker gives a prompt to wear the safety helmet irregularly.
Further, in a preferred embodiment of the present invention, the importing the second image data into the trained safety helmet wearing model to obtain a probability value includes the following steps:
judging whether a human body recognition model exists in the second image data;
if the human body recognition model exists, the human body recognition model is imported into the trained safety helmet wearing model, and a similar probability value is calculated.
FIG. 4 shows a flow chart of a method of determining a hazardous area;
further, in a preferred embodiment of the present invention, identifying a region where the second image data is located, obtaining a risk coefficient of each region, marking the region as a dangerous region when the risk coefficient is greater than a preset risk coefficient threshold, and obtaining a position of the dangerous region specifically includes the following steps:
s402, setting a danger coefficient according to the type of the second image data, marking an area where the second image data is located if the danger coefficient is equal to a preset danger coefficient, and setting the area as a behavior monitoring area;
s404, acquiring the behavior state of the behavior monitoring area within the preset time, wherein the behavior state comprises a working state and a non-working state;
and S406, if the behavior state is the working state, taking the behavior monitoring area as a dangerous area and obtaining the position of the dangerous area.
It should be noted that when the monitoring image collected by the camera within the preset time changes, the behavior state of the building apparatus is the working state; otherwise, the area is in the non-working state, and when the behavior state of the construction equipment is in the non-working state, it can be stated that the probability value that the area is an accident is small, and the area can be marked as a normal area.
FIG. 5 is a schematic diagram of a safety helmet of the intelligent safety identification method in a building site;
when a constructor wears a safety helmet and enters a construction site, the safety helmet is provided with a positioning unit 3 and a warning unit 1, the positioning unit 3 can feed back a position signal to safety intelligent equipment in the construction site, when the position of the constructor is closer to a dangerous area where a construction instrument works, the warning unit 1 can perform voice prompt such as 'entering the dangerous area', improving alertness ',' front danger ', stopping continuing to move ahead', and the like, so that the incidence rate of the phenomenon that the construction instrument hurts people caused by a visual field blind area can be reduced, the phenomenon that the construction instrument collides with the construction instrument in the construction site due to the visual field blind area can be avoided, and the safety of the construction process is improved. The communication unit 2 is a tool for communicating with remote intelligent safety identification equipment in a building site, such as real-time updated dangerous area positions, real-time positions of constructors and other information.
In addition, the present invention may further include the steps of:
acquiring the type of the second image data;
judging whether a human body image exists in the type of the second image data;
if the human body image exists, determining to acquire second image data containing the human body image to obtain third image data;
acquiring the position of the third image data in a monitoring area;
the position of the third image data within the monitored area is tracked, and the amount of displacement is determined based on the position of the third image data within the monitored area and the hazardous area.
It should be noted that, by this way, the condition of the image collected in the monitoring area can be further set, so that the whole process is more intelligent.
The method can further comprise the following steps:
acquiring a moving track of third image data in a monitoring area within preset time;
establishing and constructing a danger track map based on the movement track, constructing a danger prediction model, and importing the danger track map into the danger prediction model to obtain a trained danger prediction model;
acquiring the moving track of the third image data in the monitoring area within the next preset time, and importing the moving track of the third image data in the monitoring area within the next preset time into a trained danger prediction model to obtain a prediction result;
judging whether the prediction result is a preset prediction result or not;
if yes, an early warning signal is sent out in advance.
It should be noted that, because the visibility is low in the dust-haze weather or the fog weather, whether an accident occurs or not can be predicted by tracking the moving position of the third image data in the monitoring area, for example, whether an encounter between a truck and a constructor exists or not, whether an encounter between a truck and a construction equipment exists or not, and whether an encounter between a constructor and a construction equipment exists or not, where a danger trajectory diagram is a walking trajectory of the constructor within a preset time, and a danger prediction model is as follows:
wherein T is a risk prediction model; m is a human body recognition model which can be calculated from the object classification model; l is a displacement vector; j is the jth moment; t is the time of the human body in each monitoring area, and p is the correction displacement vector coefficient.
The prediction result is that when the current preset time is slowly close to and conforms to the dangerous track equation, the situation that the danger exists continuously forward is indicated, and an early warning signal is sent out to prompt constructors; the trained risk prediction model is a preset prediction result.
The method can further comprise the following steps:
acquiring a communication signal fed back by a safety helmet of a current constructor;
judging whether the remote control terminal can receive the communication signal within a preset time;
if yes, further judging whether the communication signal received by the remote control terminal within the preset time is interrupted;
if the communication signal is interrupted, the number of times of communication signal interruption in the preset time is obtained, and whether the number of times of interruption is larger than the preset number of times of interruption is judged;
if the area is larger than the preset threshold value, marking the area as a dangerous area and sending an early warning signal.
It should be noted that, along with the implementation of the construction engineering, when the building works indoors, an area with a poor signal may exist in the building site, at this time, a problem of a communication signal may exist between a communicator provided on the safety helmet and the remote control terminal, and before entering the area, the communication signal does not disappear at once but gradually deteriorates.
FIG. 6 illustrates a system block diagram of a security intelligent recognition device in a building site;
the second aspect of the present invention provides an intelligent identification device for security in a building site, the device includes a memory 41 and a processor 62, the memory 41 includes an intelligent identification method program for security in a building site, when the intelligent identification method program for security in a building site is executed by the processor 62, the following steps are implemented:
acquiring first image data in a building site;
constructing an object classification model, feeding back the first image data to the object classification model in real time, and performing object classification on the first image data in the object classification model to obtain second image data;
identifying the area where the second image data is located, acquiring the risk coefficient of each area, marking the area as a dangerous area when the risk coefficient is larger than a preset risk coefficient threshold value, and acquiring the position of the dangerous area;
calculating an actual danger range based on the position of the danger area;
obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and determining a danger grade based on the displacement, wherein the danger grade is divided into a low danger grade, a medium danger grade and a high danger grade, and if the danger grade is higher than the medium danger grade, immediately sending out an early warning signal.
It should be noted that, images acquired by a monitoring area are classified by using an object classification model, wherein various types of images are trained by using a neural network to obtain a trained object classification model, the classification model of the object is classified by using a shape feature classification method of the image, details of a human body contour and a building equipment contour in the image are enhanced after the image is sharpened, for example, the details are sharpened by using an ABF algorithm, a GLAS algorithm, and the like, so that the human body contour and the building equipment contour are obtained, and further, the shape contour in the image is classified, and the trained object classification model satisfies the following requirements:
wherein M is an object model value; l is a geometric characteristic coefficient and takes the value of 1; n is the number of contour points; i is the ith contour point; k is a core expression coefficient, and the highest likelihood of the color of the pigment in the contour is ensured;
pixel values within the contour;
is the average pixel value within the contour; b is the pixel brightness value of the contour point; c is a corrected brightness coefficient;
is the sum of the pixel values within the contour; z is the smoothness of the profile.
The main contour in the image is trained by utilizing the classification model, so that the image collected in real time is guided into the model to obtain a model corresponding to each image, and the real-time model and the training model are subjected to similar estimation, wherein the similar estimation meets the following requirements:
wherein S is a Gaussian variance value; p is a similar probability value; m is the value of the object model after training is finished,
is a real-time object model value;
it should be noted that the higher the probability that the calculated value of P is closer to 1 indicates that the target object is a target object type, and the closer the calculated value of P is to 0 indicates that the target object is not a target object type in the training model, where the gaussian variance value is obtained from the measurement instrument when the image is subjected to the noise reduction processing.
It should be noted that the position of the target object image can be obtained in various ways, for example, in ways of setting a camera of a positioning instrument in a monitoring area (at this time, the monitoring position of the camera is the position of the target object image), setting the positioning instrument on a helmet of a constructor, setting the positioning instrument on a construction equipment, and the like, when construction equipment (such as an excavator, a truck, a cargo platform, a forklift, and the like) occurs in the monitoring area, a safety coefficient (assigned value is 1) is added to the monitoring area at this time, which indicates that the danger probability value generated in the area is high; when no construction equipment is present in the monitored area, a safety factor (assigned as 0) is added to the area, which indicates that the probability value of danger occurrence in the area is low. The monitoring area where the construction equipment is located is updated in time no matter the construction equipment moves to any position of the construction site. Close to building apparatus more as constructor, the degree of danger in suggestion the place ahead region is high, reminds constructor in advance, and this can effectively avoided because the construction site is noisy and the incident that brings is influenced in step of field of vision blind area.
In this embodiment, calculating the actual danger range based on the position of the dangerous area specifically includes the following steps:
acquiring the effective working area range of the building equipment through a big data network;
acquiring a position coordinate value of the dangerous area;
obtaining a plurality of limit coordinate values according to the position coordinate values of the dangerous area and the effective working area range of the construction equipment;
and taking an area surrounded by the limit coordinate values as an actual dangerous range.
It should be noted that, when the object classification model obtains a construction equipment, an effective working range of the construction equipment, such as an effective area of an excavator during fixed-point working, an effective working area of a vibration pile driver, an effective working area of a concrete delivery pump, and the like, is obtained from the big data network, each type of construction machine has a certain effective working range during fixed-point working, and the working range of a limit position is taken as a dangerous range, such as the effective working range of the excavator: a cylinder surrounded by the maximum arm length of the excavator, a plane rectangular coordinate system is established with a driving shaft of a driving position (such as the maximum arm length of a working arm driving the excavator), and the rotating shaft is 0 datum point, and the limit coordinate value can be expressed as (Rsin theta, Rcos theta), wherein R is the maximum arm length of the working arm of the excavator, and theta is a circumferential angle formed between the rotating shaft and the rotating arm. Determining a position coordinate value of the dangerous area according to the position of the image shot by the monitoring camera, wherein the coordinate value can be expressed as
Where k is the kth limit coordinate value, the limit coordinate value of the actual danger range of the construction equipment can be expressed as
And the area enclosed by each limit coordinate value is the actual dangerous area. The remaining limit coordinate values can be obtained in this similar manner. Since the construction equipment is subject to change, the actual danger range follows the change.
In an embodiment, the calculating a displacement amount according to the actual risk range and the first coordinate value specifically includes:
and calculating a plurality of displacement amounts according to the actual danger range and the first coordinate value, screening out the minimum displacement amount from the plurality of displacement amounts, and taking the minimum displacement amount as the finally output displacement amount.
It should be noted that the actual displacement can be obtained by calculating the limit coordinate value of the working limit of the construction equipment and the coordinate value of the current monitoring area of the constructor; because the current actual working range of building apparatus is a circular work area or fan-shaped work area, thereby can obtain a plurality of displacement according to the calculation between coordinate and the coordinate, select one of them minimum displacement as the displacement of final output, it is when this displacement is less than the predetermined displacement, accessible alarm device (install the speaker like the safety helmet on the constructor, the speaker in the building place) suggestion constructor the place ahead has danger, constructor can in time obtain the suggestion, this can avoid leading to the incident emergence of accident because the blind area of field of vision, reduce the emergence of accident.
In this embodiment, after the step of classifying the first image data to obtain the second image data, the method further includes the following steps:
constructing a safety helmet wearing model, and obtaining a safety helmet identification model and a human body identification model from the object classification model;
guiding the safety helmet identification model and the human body identification model into a safety helmet wearing model for training to obtain a trained safety helmet wearing model;
importing the second image data into the trained safety helmet wearing model to obtain a similar probability value;
judging whether the similar probability value is smaller than a preset probability value or not;
if the value is less than the preset value, a warning that the safety helmet is not worn is sent.
It should be noted that the safety helmet model can be constructed according to the modeling mode of the object classification model, that is, the safety helmet model and the human body recognition model are integrated, and the head condition of the constructor can be collected through the camera, so that whether the constructor wears the helmet or not can be judged. On the other hand, whether the wearing is standard wearing is obtained according to the similar estimation calculation mode, when the probability value is close to 1, the wearing standard of the safety helmet of the construction personnel is indicated, when the probability value is close to 0, the head of the construction personnel is not worn or the wearing of the safety helmet is not standard, and at the moment, the safety helmet can be provided with a loudspeaker to inform the construction personnel to wear the safety helmet when entering the construction area or prompt the construction personnel to wear the safety helmet regularly. When prompting constructors to wear the safety helmet regularly, if a pressure sensor is arranged in a fastening belt in the safety helmet, when the safety helmet is not fastened, a loudspeaker gives a prompt to wear the safety helmet irregularly.
In this embodiment, importing the second image data into the trained safety helmet wearing model to obtain a probability value, specifically including the following steps:
judging whether a human body recognition model exists in the second image data;
if the human body recognition model exists, the human body recognition model is imported into the trained safety helmet wearing model, and a similar probability value is calculated.
In this embodiment, identifying a region where the second image data is located, acquiring a risk coefficient of each region, when the risk coefficient is greater than a preset risk coefficient threshold, marking the region as a risk region, and acquiring a position of the risk region specifically includes the following steps:
setting a danger coefficient according to the type of the second image data, if the danger coefficient is equal to a preset danger coefficient, marking an area where the second image data is located, and setting the area as a behavior monitoring area;
acquiring the behavior state of the behavior monitoring area within a preset time, wherein the behavior state comprises a working state and a non-working state;
and if the behavior state is the working state, taking the behavior monitoring area as a dangerous area, and acquiring the position of the dangerous area.
It should be noted that when the monitoring image collected by the camera within the preset time changes, the behavior state of the building apparatus is the working state; otherwise, the area is in the non-working state, and when the behavior state of the construction equipment is in the non-working state, it can be stated that the probability value that the area is an accident is small, and the area can be marked as a normal area.
Fig. 7 shows a block diagram of a security intelligent recognition device in a building site.
A third aspect of the present invention provides an intelligent identification device for security in a building site, including:
the acquisition module 10 is used for acquiring first image data in a building site;
the classification module 20 is configured to construct an object classification model, feed back the first image data to the object classification model in real time, and perform object classification on the first image data in the object classification model to obtain second image data;
the identification module 30 is configured to identify a region where the second image data is located, acquire a risk coefficient of each region, mark the region as a risk region when the risk coefficient is greater than a preset risk coefficient threshold, and acquire a position of the risk region;
a first calculation module 40, which calculates an actual danger range based on the position of the danger area;
the second calculation module 50 is used for obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and the early warning module 60 determines the danger level based on the displacement, wherein the danger level is divided into a low danger level, a medium danger level and a high danger level, and if the danger level is higher than the medium danger level, an early warning signal is immediately sent out.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only one logical function division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; either at one location or distributed across multiple network elements; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Alternatively, the integrated unit of the present invention may be stored in a computer-readable storage medium if it is implemented in the form of a software functional module and sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or a part contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods of the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, a ROM, a RAM, a magnetic or optical disk, or various other media that can store program code.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. An intelligent identification method for safety in a building site is characterized by comprising the following steps:
collecting first image data in a building site;
constructing an object classification model, feeding back the first image data to the object classification model in real time, and performing object classification on the first image data in the object classification model to obtain second image data;
identifying the area where the second image data is located, acquiring the risk coefficient of each area, marking the area as a dangerous area when the risk coefficient is larger than a preset risk coefficient threshold value, and acquiring the position of the dangerous area;
calculating an actual danger range based on the position of the danger area;
obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and determining a danger grade based on the displacement, wherein the danger grade is divided into a low danger grade, a medium danger grade and a high danger grade, and if the danger grade is higher than the medium danger grade, immediately sending out an early warning signal.
2. The intelligent identification method for safety in a building site according to claim 1, wherein an actual danger range is calculated based on the position of the dangerous area, and the method specifically comprises the following steps:
acquiring the effective working area range of the building equipment through a big data network;
acquiring a position coordinate value of the dangerous area;
obtaining a plurality of limit coordinate values according to the position coordinate values of the dangerous area and the effective working area range of the construction equipment;
and taking an area surrounded by the limit coordinate values as an actual dangerous range.
3. The method according to claim 1, wherein the step of obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual risk range and the first coordinate value specifically comprises:
and calculating a plurality of displacement amounts according to the actual danger range and the first coordinate value, screening out the minimum displacement amount from the plurality of displacement amounts, and taking the minimum displacement amount as the finally output displacement amount.
4. The intelligent identification method for safety in a building site according to claim 1, wherein an object classification model is constructed, the first image data is fed back to the object classification model in real time, and in the object classification model, object classification is performed on the first image data, and the step of obtaining second image data further comprises the following steps:
constructing a safety helmet wearing model, and obtaining a safety helmet identification model and a human body identification model from the object classification model;
guiding the safety helmet identification model and the human body identification model into a safety helmet wearing model for training to obtain a trained safety helmet wearing model;
importing the second image data into the trained safety helmet wearing model to obtain a similar probability value;
judging whether the similar probability value is smaller than a preset probability value or not;
if the value is less than the preset value, a warning that the safety helmet is not worn is sent.
5. The intelligent identification method for safety in a building site according to claim 4, wherein the second image data is imported into the trained safety helmet wearing model to obtain a probability value, and the method specifically comprises the following steps:
judging whether a human body recognition model exists in the second image data or not;
if the human body recognition model exists, the human body recognition model is imported into the trained safety helmet wearing model, and a similar probability value is calculated.
6. The intelligent identification method for safety in a building site according to claim 1, wherein an area where the second image data is located is identified, a risk coefficient of each area is obtained, when the risk coefficient is greater than a preset risk coefficient threshold, the area is marked as a dangerous area, and a position of the dangerous area is obtained, and specifically, the method comprises the following steps:
setting a danger coefficient according to the type of the second image data, if the danger coefficient is equal to a preset danger coefficient, marking an area where the second image data is located, and setting the area as a behavior monitoring area;
acquiring the behavior state of the behavior monitoring area within preset time, wherein the behavior state comprises a working state and a non-working state;
and if the behavior state is the working state, taking the behavior monitoring area as a dangerous area, and acquiring the position of the dangerous area.
7. An intelligent identification device for safety in a building site is characterized by comprising a memory and a processor, wherein the memory comprises an intelligent identification method program for safety in the building site, and when the intelligent identification method program for safety in the building site is executed by the processor, the following steps are realized:
collecting first image data in a building site;
constructing an object classification model, feeding back the first image data to the object classification model in real time, and performing object classification on the first image data in the object classification model to obtain second image data;
identifying the area where the second image data is located, acquiring the risk coefficient of each area, marking the area as a dangerous area when the risk coefficient is larger than a preset risk coefficient threshold value, and acquiring the position of the dangerous area;
calculating an actual danger range based on the position of the danger area;
obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and determining danger levels based on the displacement, wherein the danger levels are divided into a low danger level, a medium danger level and a high danger level, and if the danger level is higher than the medium danger level, immediately sending out an early warning signal.
8. The intelligent identification device for safety in a building site according to claim 7, wherein the actual danger range is calculated based on the position of the dangerous area, specifically comprising the following steps:
acquiring the effective working area range of the building equipment through a big data network;
acquiring a position coordinate value of the dangerous area;
obtaining a plurality of limit coordinate values according to the position coordinate values of the dangerous area and the effective working area range of the construction equipment;
and taking an area surrounded by the limit coordinate values as an actual dangerous range.
9. The intelligent identification device for safety in a building site according to claim 7, wherein an object classification model is constructed and the first image data is fed back to the object classification model in real time, and in the object classification model, the step of performing object classification on the first image data to obtain the second image data further comprises the following steps:
constructing a safety helmet wearing model, and obtaining a safety helmet identification model and a human body identification model from the object classification model;
guiding the safety helmet identification model and the human body identification model into a safety helmet wearing model for training to obtain a trained safety helmet wearing model;
importing the second image data into the trained safety helmet wearing model to obtain a similar probability value;
judging whether the similar probability value is smaller than a preset probability value or not;
if the value is less than the preset value, a warning that the safety helmet is not worn is sent.
10. The utility model provides a security intelligent recognition equipment in building site which characterized in that, security intelligent recognition equipment includes in the building site:
the acquisition module acquires first image data in a building site;
the classification module is used for constructing an object classification model, feeding back the first image data to the object classification model in real time, and performing object classification on the first image data in the object classification model to obtain second image data;
the identification module is used for identifying the area where the second image data is located, acquiring the danger coefficient of each area, marking the area as a dangerous area when the danger coefficient is larger than a preset danger coefficient threshold value, and acquiring the position of the dangerous area;
the first calculation module is used for calculating an actual danger range based on the position of the danger area;
the second calculation module is used for obtaining the real-time coordinate position of the human body image from the second image data to obtain a first coordinate value, and calculating the displacement according to the actual danger range and the first coordinate value;
and the early warning module is used for determining a danger grade based on the displacement, wherein the danger grade is divided into a low danger grade, a medium danger grade and a high danger grade, and if the danger grade is higher than the medium danger grade, an early warning signal is sent out immediately.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210511276.0A CN114612866B (en) | 2022-05-12 | 2022-05-12 | Intelligent identification method, device and equipment for safety in building site |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210511276.0A CN114612866B (en) | 2022-05-12 | 2022-05-12 | Intelligent identification method, device and equipment for safety in building site |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114612866A true CN114612866A (en) | 2022-06-10 |
| CN114612866B CN114612866B (en) | 2022-09-02 |
Family
ID=81870511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210511276.0A Active CN114612866B (en) | 2022-05-12 | 2022-05-12 | Intelligent identification method, device and equipment for safety in building site |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114612866B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114783144A (en) * | 2022-06-17 | 2022-07-22 | 深圳市易智博网络科技有限公司 | Intelligent building site safety monitoring method and device and computer equipment |
| CN115860443A (en) * | 2023-03-02 | 2023-03-28 | 青岛研博数据信息技术有限公司 | Building intelligent management method and system based on BIM technology |
| CN116246421A (en) * | 2023-05-04 | 2023-06-09 | 山东领创信息科技有限公司 | Safety early warning method, system and equipment based on intelligent wearing |
| CN116895043A (en) * | 2023-06-13 | 2023-10-17 | 郑州宝冶钢结构有限公司 | Intelligent safety monitoring and early warning method, system and storage medium for construction site |
| CN117132947A (en) * | 2023-10-27 | 2023-11-28 | 山东拓新电气有限公司 | Dangerous area personnel identification method based on monitoring video |
| CN118710139A (en) * | 2024-08-29 | 2024-09-27 | 国网山东省电力公司营销服务中心(计量中心) | Electric power project field management and control supervision method and system based on AI technology |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110232320A (en) * | 2019-05-08 | 2019-09-13 | 华中科技大学 | Method and system of the real-time detection building-site worker danger close to construction machinery |
| CN110443976A (en) * | 2019-08-14 | 2019-11-12 | 深圳市沃特沃德股份有限公司 | Safety prompt function method, apparatus and storage medium based on safety cap |
| CN110745704A (en) * | 2019-12-20 | 2020-02-04 | 广东博智林机器人有限公司 | Tower crane early warning method and device |
| US20200150656A1 (en) * | 2018-11-12 | 2020-05-14 | Alberto Daniel Lacaze | Autonomous Trucks with Specialized Behaviors for Mining and Construction Applications |
| CN111392619A (en) * | 2020-03-25 | 2020-07-10 | 广东博智林机器人有限公司 | Tower crane early warning method, device and system and storage medium |
| CN111784967A (en) * | 2020-07-23 | 2020-10-16 | 南通市达欣工程股份有限公司 | Building construction safety early warning protection system based on BIM |
| CN112784821A (en) * | 2021-03-06 | 2021-05-11 | 深圳市安比智慧科技有限公司 | Building site behavior safety detection and identification method and system based on YOLOv5 |
| CN112862887A (en) * | 2020-12-31 | 2021-05-28 | 卓喜龙 | Building construction safety management early warning system and safety cloud platform based on artificial intelligence and big data analysis |
| CN113536935A (en) * | 2021-06-17 | 2021-10-22 | 深圳有象智联科技有限公司 | Safety monitoring method and equipment for engineering site |
| CN113762229A (en) * | 2021-11-10 | 2021-12-07 | 山东天亚达新材料科技有限公司 | Intelligent identification method and system for building equipment in building site |
| CN113808354A (en) * | 2021-11-16 | 2021-12-17 | 深圳市思拓通信系统有限公司 | Method, device and medium for early warning of construction site dangerous area |
| CN114209118A (en) * | 2021-12-29 | 2022-03-22 | 国网瑞嘉(天津)智能机器人有限公司 | High-altitude operation intelligent early warning method and device and intelligent safety helmet |
-
2022
- 2022-05-12 CN CN202210511276.0A patent/CN114612866B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200150656A1 (en) * | 2018-11-12 | 2020-05-14 | Alberto Daniel Lacaze | Autonomous Trucks with Specialized Behaviors for Mining and Construction Applications |
| CN110232320A (en) * | 2019-05-08 | 2019-09-13 | 华中科技大学 | Method and system of the real-time detection building-site worker danger close to construction machinery |
| CN110443976A (en) * | 2019-08-14 | 2019-11-12 | 深圳市沃特沃德股份有限公司 | Safety prompt function method, apparatus and storage medium based on safety cap |
| CN110745704A (en) * | 2019-12-20 | 2020-02-04 | 广东博智林机器人有限公司 | Tower crane early warning method and device |
| CN111392619A (en) * | 2020-03-25 | 2020-07-10 | 广东博智林机器人有限公司 | Tower crane early warning method, device and system and storage medium |
| CN111784967A (en) * | 2020-07-23 | 2020-10-16 | 南通市达欣工程股份有限公司 | Building construction safety early warning protection system based on BIM |
| CN112862887A (en) * | 2020-12-31 | 2021-05-28 | 卓喜龙 | Building construction safety management early warning system and safety cloud platform based on artificial intelligence and big data analysis |
| CN112784821A (en) * | 2021-03-06 | 2021-05-11 | 深圳市安比智慧科技有限公司 | Building site behavior safety detection and identification method and system based on YOLOv5 |
| CN113536935A (en) * | 2021-06-17 | 2021-10-22 | 深圳有象智联科技有限公司 | Safety monitoring method and equipment for engineering site |
| CN113762229A (en) * | 2021-11-10 | 2021-12-07 | 山东天亚达新材料科技有限公司 | Intelligent identification method and system for building equipment in building site |
| CN113808354A (en) * | 2021-11-16 | 2021-12-17 | 深圳市思拓通信系统有限公司 | Method, device and medium for early warning of construction site dangerous area |
| CN114209118A (en) * | 2021-12-29 | 2022-03-22 | 国网瑞嘉(天津)智能机器人有限公司 | High-altitude operation intelligent early warning method and device and intelligent safety helmet |
Non-Patent Citations (2)
| Title |
|---|
| Y. ZHAO 等: "Deep Learning for Risk Detection and Trajectory Tracking at Construction Sites", 《IEEE ACCESS》 * |
| 马辉 等: "基于空间单元分析的装配式建筑施工安全预警模型", 《中国安全科学学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114783144A (en) * | 2022-06-17 | 2022-07-22 | 深圳市易智博网络科技有限公司 | Intelligent building site safety monitoring method and device and computer equipment |
| CN114783144B (en) * | 2022-06-17 | 2022-09-06 | 深圳市易智博网络科技有限公司 | Intelligent building site safety monitoring method and device and computer equipment |
| CN115860443A (en) * | 2023-03-02 | 2023-03-28 | 青岛研博数据信息技术有限公司 | Building intelligent management method and system based on BIM technology |
| CN115860443B (en) * | 2023-03-02 | 2023-06-02 | 青岛研博数据信息技术有限公司 | Building intelligent management method and system based on BIM technology |
| CN116246421A (en) * | 2023-05-04 | 2023-06-09 | 山东领创信息科技有限公司 | Safety early warning method, system and equipment based on intelligent wearing |
| CN116895043A (en) * | 2023-06-13 | 2023-10-17 | 郑州宝冶钢结构有限公司 | Intelligent safety monitoring and early warning method, system and storage medium for construction site |
| CN116895043B (en) * | 2023-06-13 | 2024-01-26 | 郑州宝冶钢结构有限公司 | Intelligent safety monitoring and early warning method, system and storage medium for construction site |
| CN117132947A (en) * | 2023-10-27 | 2023-11-28 | 山东拓新电气有限公司 | Dangerous area personnel identification method based on monitoring video |
| CN117132947B (en) * | 2023-10-27 | 2024-01-26 | 山东拓新电气有限公司 | Dangerous area personnel identification method based on monitoring video |
| CN118710139A (en) * | 2024-08-29 | 2024-09-27 | 国网山东省电力公司营销服务中心(计量中心) | Electric power project field management and control supervision method and system based on AI technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114612866B (en) | 2022-09-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114612866B (en) | Intelligent identification method, device and equipment for safety in building site | |
| CN111275923B (en) | Man-machine collision early warning method and system for construction site | |
| CN106341661B (en) | Patrol robot | |
| CN111553214B (en) | Method and system for detecting smoking behavior of driver | |
| CN114885133A (en) | Depth image-based equipment safety real-time monitoring method and system and related equipment | |
| CN111931706B (en) | Man-machine collision early warning method and system for construction site | |
| CN112862887A (en) | Building construction safety management early warning system and safety cloud platform based on artificial intelligence and big data analysis | |
| CN113469115B (en) | Method and device for outputting information | |
| Pech et al. | Head tracking based glance area estimation for driver behaviour modelling during lane change execution | |
| CN114782897A (en) | Dangerous behavior detection method and system based on machine vision and deep learning | |
| CN112464782A (en) | Pedestrian identification method and system | |
| CN117474321B (en) | BIM model-based construction site risk intelligent identification method and system | |
| CN117523537A (en) | Dynamic judging method for dangerous degree of vehicle driving | |
| WO2022244747A1 (en) | Workplace management system, workplace management method, and workplace management program | |
| CN113179389A (en) | System and method for identifying crane jib of power transmission line dangerous vehicle | |
| CN116259002A (en) | Human body dangerous behavior analysis method based on video | |
| CN103834796A (en) | Method for online recognition of deviation of strip steel in furnace | |
| CN114445398A (en) | Method and device for monitoring state of side protection plate of hydraulic support of coal mining machine | |
| KR102541212B1 (en) | Embeded image recognition safety monitering and controling integrated platform comprising ai based image recognition system | |
| CN115050125B (en) | 2d camera-based safety early warning method, device, equipment and storage medium | |
| CN115100495B (en) | Light-weight safety helmet detection method based on sub-feature fusion | |
| CN116703157A (en) | Remote visual display method and device for construction site, electronic equipment and medium | |
| CN112232136B (en) | Vehicle safety belt detection method and device, electronic equipment and storage medium | |
| CN115661756B (en) | Urban accident monitoring method and device based on guardrails and network side server | |
| CN116977920B (en) | Critical protection method for multi-zone type multi-reasoning early warning mechanism |
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 |