Method, medium and system for dynamically monitoring safety of labor personnel on construction site

Abstract

The invention discloses a method, a medium and a system for dynamically monitoring the safety of labor service personnel on a construction site, the system comprises a monitoring platform, the monitoring platform is provided with a control host, the control host is connected with an acquisition module, and the control host and the acquisition module are connected through radio signals or wired electric signals; the control host is provided with a display screen, the control host is in wireless signal connection with a mobile terminal, and the control host is further provided with an alarm module, a storage module and a data analysis module; the acquisition module is used for acquiring field parameter information of constructors; the data analysis module is used for analyzing and processing the parameters acquired by the acquisition module; obtaining a target of the constructor, identifying the target, extracting safety features, and judging the safety condition of the constructor according to the safety features; the safety production accident cost is reduced, and the direct economic loss is saved. The construction efficiency is improved and the construction period is shortened. The enterprise safety management level is improved, and the safety input cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of safety monitoring, and in particular relates to a method, a medium and a system for dynamically monitoring the safety of laborers at a construction site. Background Art

[0002] Comply with the requirements of production safety laws and regulations. National production safety laws and regulations clearly require the establishment of a sound production safety supervision and management system and the adoption of necessary measures to prevent and control occupational risks. Prevent major safety accidents. Accidents frequently occur at construction sites. Dynamic monitoring can promptly detect hidden dangers and abnormalities, effectively prevent accidents, and reduce casualties and economic losses. Improve safety management level. Dynamic monitoring can grasp the safety status in real time, provide a basis for the discovery and resolution of problems, optimize the safety management process, and improve the efficiency and level of supervision.

[0003] Traditional video surveillance provides real-time surveillance images to staff, who then judge the abnormal status of personnel and whether there is intrusion. This manual processing method will cause fatigue to managers over a long period of time, and it is inefficient and cannot handle security issues in a timely manner. At the same time, this method only has the monitoring process but no prevention and control, and cannot fundamentally solve the key issues of security.

[0004] At present, video surveillance is developing in the direction of intelligence. Intelligent video surveillance systems use image processing technology and deep learning to detect people in video images and judge intrusions, replacing the original manual monitoring. Computers will not get tired due to long-term work. Compared with traditional video surveillance, it is more stable and has more functions. However, construction sites are complex and diverse, and a large number of mechanical equipment obstructions and dust in the environment will have a certain impact on the detection effect of video surveillance personnel, and the recognition accuracy needs to be improved; and in terms of the safety monitoring of laborers, a construction site monitoring system is urgently needed to be responsible for the safety of laborers. Summary of the invention

[0005] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a method, medium and system for dynamically monitoring the safety of laborers on a construction site, thereby solving at least one of the problems of the background technology.

[0006] The present invention provides the following technical solutions:

[0007] A system for dynamically monitoring the safety of laborers at a construction site includes a monitoring platform, the monitoring platform is provided with a control host, the control host is connected to a collection module, and the connection and transmission mode between the control host and the collection module adopts a radio signal or a wired electric signal connection;

[0008] The control host is provided with a display screen, the control host wireless signal is connected to the mobile terminal, and the control host is also provided with an alarm module, a storage module, and a data analysis module;

[0009] The collection module is used to collect on-site parameter information of construction personnel;

[0010] The data analysis module is used to analyze and process the parameters collected by the acquisition module; obtain the construction personnel targets, identify the targets, extract safety features, and judge the safety status of the construction personnel based on the safety features.

[0011] Preferably, the acquisition module includes a plurality of cameras, each camera is electrically connected to a video optical terminal, the video optical terminal is electrically connected to an encoder, and the encoder is connected to a control host of the monitoring platform via a wireless or wired network.

[0012] Preferably, the acquisition module also includes a body temperature measurement sensor and a heart rate and blood oxygen sensor. The body temperature measurement sensor and heart rate and blood oxygen sensor acquisition module is integrated with an A / D converter, which is converted into a digital signal through the A / D converter. The acquisition module transmits the signal to the control host through the integrated LoRa wireless communication mode, and the control host receives the physiological signal. The acquisition module is also integrated with a GPS positioning function.

[0013] Preferably, the LoRa wireless communication module is an ATK-LORA-01 wireless serial port module, which is powered by 5V. The module has an operating frequency of 410-441MHz, with a step of 1MHz, a total of 32 channels, multiple working modes, and an SMA antenna. The maximum transmission distance is 3km; the module TXD port is connected to the PB11 port of the main control chip, the RXD port is connected to the PB10 port of the main control chip, the AUX port is connected to the PA4 port of the main control chip, and the MDO port is connected to the PA15 port of the main control chip; directional transmission is used, and the target node address and channel are set to transmit the collected vital sign data.

[0014] Preferably, the alarm module of the control host includes a buzzer, which receives instructions from the control host; the storage module stores the collected information in the control host.

[0015] A readable storage medium for dynamically monitoring the safety of laborers at a construction site, wherein control instructions are stored in the readable storage medium, and the readable storage medium is used in a system for dynamically monitoring the safety of laborers at a construction site as described in any one of claims 1-5.

[0016] A method for dynamically monitoring the safety of laborers at a construction site, the method comprising: step 1: information collection, the information collection comprising image collection and physiological characteristic collection, the image collection is performed on the construction site through a camera, and the physiological characteristic collection is performed through a body temperature detection sensor and a heart rate and blood oxygen sensor to collect the body temperature, heart rate and blood oxygen concentration of the laborer; the body temperature detection sensor is arranged inside the laborer's helmet to collect the temperature of the laborer, and the heart rate and blood oxygen sensor is arranged inside the laborer's safety belt near the heart;

[0017] Step 2: The collected temperature data and heart rate oxygen data of the laborers are transmitted to the control host, which processes the collected data and makes a judgment. When the set temperature and heart rate oxygen data thresholds are reached, the control host sends a command to the alarm module to sound an alarm.

[0018] Step 3: After obtaining the construction site image through the camera, the control center pre-processes the image, performs moving target detection, target recognition, and safety detection. If the detection data result shows that the laborers do not wear safety belts and helmets properly, a command is sent to the alarm module and the buzzer issues an early warning.

[0019] Preferably, after the image information is collected and the control center obtains the on-site image information, it first determines whether a danger zone has been set for the obtained image. If a danger zone has been set for the work site, the control center saves a template image with the danger zone set. After acquiring the image, the acquired image information is compared with the saved template image of the danger zone to perform moving target recognition and safety detection on the labor personnel in the danger zone. If no danger zone has been set for the work site, moving target recognition and safety detection are performed on the labor personnel in the entire image.

[0020] Preferably, the setting of the danger zone is to obtain complete picture information of a construction site, and in the process of processing the picture, obtain data of four boundary fixed points of the danger zone of the image, obtain the coordinates of the four points, and connect the four coordinate points into a quadrilateral area, which is the danger zone.

[0021] Preferably, in step 2, the temperature and heart rate blood oxygen sensor data are collected every 200ms, and the control center obtains the location information of the service personnel through the GPS function. The GPS collects the location information every 5s and transmits it to the control center; the control center saves the sensor data collected in the last two time intervals in the corresponding array, and eliminates abnormal data through data preprocessing; after the data collection of this time interval is completed, the serial port interrupt is disabled, the reception of sensor data is suspended, and the sensor data in this time interval is obtained as the final transmission data, and the data is displayed on the display screen and the mobile terminal.

[0022] Preferably, the body temperature detection sensor is a GY614V3 non-contact infrared temperature measurement sensor, which is powered by 3 to 5V. The initial communication is in serial port UART mode, and it automatically outputs data frames, 1-byte address, 1-byte function code, byte data (including target temperature and ambient temperature) and checksum; by connecting to serial port 5 of the control host chip, the module TD port transmits data frames to the control host chip PD2 port according to the set module update rate, without the need for AT command query.

[0023] In addition, the heart rate blood oxygen sensor MAX30102 heart rate sensor is powered by 5V and embedded with STM32F0 low-power chip for data processing and communication mode conversion. 2 C communication is converted into serial communication to output heart rate and blood oxygen data; the TX port of the heart rate and blood oxygen sensor module is connected to the PC11 port of the control host chip, and the RX port is connected to the PC10 port of the control host chip. The heart rate and blood oxygen sensor module sends AT instructions through the control host chip and passively outputs the sensor data to be queried.

[0024] In addition, the GPS positioning module is the ATK-S1216F8_BD Beidou / GPS dual-mode positioning module, which is powered by 3.3V / 5V. The module has its own IPX interface and can be connected to an active antenna. The module's TXD port is connected to the PA3 port of the main control chip, the RXD port is connected to the PA2 port of the main control chip, and the PPS port outputs a standard clock pulse, which is not used in this system, and the corresponding pin is floating. Through the set GPS positioning module, when the physiological parameters of the laborers are abnormal and an early warning is issued through the buzzer, they can be located in time, facilitating timely rescue.

[0025] The server of the monitoring platform sets up the TCP / IP communication protocol through the Socket socket. The server monitors the communication, and the host computer of the gateway node makes a communication request. If the communication link between the two is successfully established, it will start to receive data. At this time, the communication requests transmitted by other gateway nodes will enter the communication waiting queue. The gateway node forwards the vital sign data and address data collected by each terminal node for its summary. By identifying the node number, the corresponding data is updated, and then the vital sign data corresponding to each terminal node is detected for abnormalities through the control center. If it is within the safe range, only the vital sign information is displayed on the display. If abnormal vital sign data appears, a safety warning is sent. After the gateway node has received the data, the current communication connection is terminated.

[0026] When obtaining body temperature and heart rate and blood oxygen concentration through body temperature detection sensors and heart rate and blood oxygen sensors, due to individual deviations, these deviations will often give false alarms, which not only affects the accuracy of the early warning, but also wastes manpower and material resources, affecting the judgment of the safety monitoring of laborers. In order to further increase the accuracy of monitoring the vital signs of laborers, the acquired vital signs data of laborers are processed, abnormal data are eliminated, and the accuracy of data monitoring is increased. The process of processing the acquired data is that the acquired multiple vital signs collection data is x=(x 1 ,x 2 ,...,x m ), μ is the mean of x, σ is the standard deviation, μ=(x 1 +x 2 +...+x m ) / m;σ=(1 / (n-1)(∑i x i 2 -∑ i μ i 2 / n)) 1 / 2 ; n is the number of data values ​​obtained; then when |x j -μ|≥3σ, the obtained x j The data is abnormal data; the abnormal data will be eliminated to improve the accuracy of the acquired data.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The method, medium and system for dynamically monitoring the safety of laborers at construction sites of the present invention can reduce the cost of production safety accidents and save direct economic losses. It can improve construction efficiency and shorten the construction period. It can improve the safety management level of enterprises and reduce the cost of safety investment. It can stimulate the development of the safety monitoring industry and drive related industries; improve the level of production safety at construction sites and protect the rights and interests of workers. It can reduce the occurrence of safety accidents and reduce social risks. It can enhance the public's awareness of production safety. It can improve the safety supervision system and improve the supervision capacity; reduce the risk of accidents and the risk of environmental pollution through safety monitoring. New technologies promote resource conservation and environmentally friendly production. The construction of a production safety system promotes green construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0030] Figure 1 It is a system block diagram of the present invention.

[0031] Figure 2 It is a block diagram of the control host of the present invention.

[0032] Figure 3 It is a flow chart of the monitoring method of the present invention.

[0033] Figure 4 It is a flow chart of the early warning method of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] In addition, in the present invention, descriptions such as "first" and "second" are only used for descriptive purposes, and do not specifically refer to the order or sequence, nor are they used to limit the present invention. They are only used to distinguish components or operations described with the same technical terms, and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions and technical features between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0037] Embodiment 1:

[0038] refer to Figure 1-4 A system for dynamically monitoring the safety of laborers on construction sites includes a monitoring platform. The monitoring platform is provided with a control host, and the control host is connected to a collection module. The connection transmission method between the control host and the collection module adopts a radio signal or a wired signal connection; the control host is provided with a display screen, and the control host is wirelessly connected to a mobile terminal. The control host is also provided with an alarm module, a storage module, and a data analysis module; the collection module is used to collect on-site parameter information of construction personnel; the data analysis module is used to analyze and process the parameters collected by the collection module; the construction personnel targets are obtained, the targets are identified, the safety features are extracted, and the safety situation of the construction personnel is judged according to the safety features.

[0039] The acquisition module includes multiple cameras, each of which is electrically connected to a video optical terminal, which is electrically connected to an encoder, and the encoder is connected to the control host of the monitoring platform via a wireless or wired network. The acquisition module also includes a body temperature measurement sensor and a heart rate and blood oxygen sensor. The body temperature measurement sensor and the heart rate and blood oxygen sensor acquisition module are integrated with an A / D converter, which is converted into a digital signal through the A / D converter. The acquisition module transmits the signal to the control host through the integrated LoRa wireless communication mode, and the control host receives the physiological signal; the acquisition module is also integrated with a GPS positioning function.

[0040] The LoRa wireless communication module is the ATK-LORA-01 wireless serial port module, which is powered by 5V. The module operates at a frequency of 410-441MHz, in 1MHz increments, with a total of 32 channels. It has multiple working modes and uses an SMA antenna with a maximum transmission distance of 3km. The module's TXD port is connected to the PB11 port of the main control chip, the RXD port is connected to the PB10 port of the main control chip, the AUX port is connected to the PA4 port of the main control chip, and the MDO port is connected to the PA15 port of the main control chip. Directional transmission is used to set the target node address and channel to transmit the collected vital sign data.

[0041] The alarm module of the control host includes a buzzer, which receives the command of the control host; the storage module stores the collected information in the control host. A readable storage medium for dynamically monitoring the safety of laborers at a construction site, wherein control commands are stored in the readable storage medium, and the readable storage medium is used in a system for dynamically monitoring the safety of laborers at a construction site as described in any one of claims 1 to 5.

[0042] A method for dynamically monitoring the safety of laborers at a construction site, the method comprising: step 1: information collection, the information collection comprising image collection and physiological characteristic collection, the image collection is performed on the construction site through a camera, and the physiological characteristic collection is performed through a body temperature detection sensor and a heart rate and blood oxygen sensor to collect the body temperature, heart rate and blood oxygen concentration of the laborer; the body temperature detection sensor is arranged inside the laborer's helmet to collect the temperature of the laborer, and the heart rate and blood oxygen sensor is arranged inside the laborer's safety belt near the heart;

[0043] Step 2: The collected temperature data and heart rate oxygen data of the laborers are transmitted to the control host, which processes the collected data and makes a judgment. When the set temperature and heart rate oxygen data thresholds are reached, the control host sends a command to the alarm module to sound an alarm.

[0044] Step 3: After obtaining the construction site image through the camera, the control center pre-processes the image, performs moving target detection, target recognition, and safety detection. If the detection data result shows that the laborers do not wear safety belts and helmets properly, a command is sent to the alarm module and the buzzer issues an early warning.

[0045] After the image information is collected and the control center obtains the on-site image information, it first determines whether a dangerous area is set for the acquired image. If a dangerous area is set for the work site, the control center saves the template image of the dangerous area. After acquiring the image, it compares the acquired image information with the saved template image of the dangerous area, and performs moving target recognition and safety detection on the laborers in the dangerous area; if the work site does not set a dangerous area, the laborers in the entire image are subjected to moving target recognition and safety detection. The setting of the dangerous area is to obtain a complete picture information of the construction site. In the process of processing the picture, the data of the four boundary fixed points of the image dangerous area are obtained, and the coordinates of the four points are obtained. These four coordinate points are connected into a quadrilateral area, which is the dangerous area.

[0046] In step 2, the temperature, heart rate and blood oxygen sensor data are collected every 200ms, and the control center obtains the location information of the labor personnel through the GPS function. The GPS collects the location information every 5s and transmits it to the control center; the control center saves the sensor data collected in the last two time intervals in the corresponding array, and removes abnormal data through data preprocessing; after the data collection of this time interval is completed, the serial port interrupt is disabled, the reception of sensor data is suspended, and the sensor data in this time interval is obtained as the final transmission data, and the data is displayed on the display screen and the mobile terminal.

[0047] The body temperature detection sensor is a GY614V3 non-contact infrared temperature sensor, powered by 3~5V. The initial communication is in serial UART mode, automatically outputting data frames, 1-byte address, 1-byte function code, byte data (including target temperature and ambient temperature) and checksum; by connecting to serial port 5 of the control host chip, the module TD port transmits data frames to the control host chip PD2 port according to the set module update rate, without the need for AT command query.

[0048] The heart rate blood oxygen sensor MAX30102 heart rate sensor is powered by 5V and embedded with STM32F0 low-power chip for data processing and communication mode conversion. 2 C communication is converted into serial communication to output heart rate and blood oxygen data; the TX port of the heart rate and blood oxygen sensor module is connected to the PC11 port of the control host chip, and the RX port is connected to the PC10 port of the control host chip. The heart rate and blood oxygen sensor module sends AT instructions through the control host chip and passively outputs the sensor data to be queried.

[0049] The GPS positioning module is the ATK-S1216F8_BD Beidou / GPS dual-mode positioning module, which is powered by 3.3V / 5V. The module has its own IPX interface and can be connected to an active antenna. The module's TXD port is connected to the PA3 port of the main control chip, the RXD port is connected to the PA2 port of the main control chip, and the PPS port outputs a standard clock pulse, which is not used in this system, and the corresponding pin is floating. Through the set GPS positioning module, when the physiological parameters of the laborers are abnormal and an early warning is issued through the buzzer, they can be located in time, facilitating timely rescue.

[0050] Embodiment 2:

[0051] On the basis of Example 1, the server of the monitoring platform sets the TCP / IP communication protocol through the Socket socket. The server side monitors the communication, and the host computer side of the gateway node makes a communication request. If the communication link connection between the two is successfully established, data will be received. At this time, the communication requests transmitted by other gateway nodes will enter the communication waiting queue. The gateway node forwards the vital sign data and address data collected by each terminal node for its summary. By identifying the node number, the corresponding data is updated, and then the vital sign data corresponding to each terminal node is detected for abnormality through the control center. If it is within the safe range, only the vital sign information is displayed through the display. If abnormal vital sign data appears, a safety warning is sent. After the gateway node data is received, the current communication connection is terminated.

[0052] When obtaining body temperature and heart rate and blood oxygen concentration through body temperature detection sensors and heart rate and blood oxygen sensors, due to individual deviations, these deviations will often give false alarms, which not only affects the accuracy of the early warning, but also wastes manpower and material resources, affecting the judgment of the safety monitoring of laborers. In order to further increase the accuracy of monitoring the vital signs of laborers, the acquired vital signs data of laborers are processed, abnormal data are eliminated, and the accuracy of data monitoring is increased. The process of processing the acquired data is that the acquired multiple vital signs collection data is x=(x 1 ,x 2 ,...,x m ), μ is the mean of x, σ is the standard deviation, μ=(x 1 +x 2 +...+x m ) / m;σ=(1 / (n-1)(∑ i x i 2 -∑ i μ i 2 / n)) 1 / 2 ; n is the number of data values ​​obtained; then when |x j -μ|≥3σ, the obtained x jThe data is abnormal data; the abnormal data will be eliminated to improve the accuracy of the acquired data.

[0053] The process of detecting moving targets of construction site personnel is as follows: the first step is to grayscale the acquired images of on-site construction labor personnel, and then extract the construction personnel through morphological processing; the second step is to obtain the target personnel image after morphological processing, and only extract features from the binary area within the area of ​​the circumscribed rectangle of the acquired personnel image, and filter out the interference information of the small area within the circumscribed rectangle; the third step is to finally cut out the personnel target in the image, and process the cutout image into 64*64, and finally perform classification feature extraction on the image information, and complete the identification of the moving target personnel according to the classification feature extraction. The specific process of the first step of extracting the target of the construction personnel is as follows: the area ratio A and the perimeter ratio S of the object in the picture are used as two features for target recognition, then A satisfies, A=A1 / A2; S satisfies, S=S1 / S2; in the above formula, A1 is the area of ​​the target area of ​​the personnel, A2 is the area of ​​the circumscribed rectangle of the target area, S1 is the perimeter of the target area of ​​the personnel, and S2 is the perimeter of the circumscribed rectangle of the target area; the parameters of the area and perimeter are described by the number of pixels; since the perimeter and area of ​​the human body area are larger than the area and perimeter of its outer rectangle, for the construction personnel The area and perimeter of the tool cart and work cart area are slightly different from the area and perimeter of their circumscribed rectangles; and when a person walks upright, the height-to-width ratio of the human body is large, while the height-to-width ratio of the tool cart and work cart is small, and the height-to-width ratio R of the human body area satisfies R=H / W; H is the height of the target circumscribed rectangle, and W is the width of the target circumscribed rectangle; the height and width parameters are described by the number of pixels; so when the perimeter and area of ​​the target area in the acquired image are larger than the area and perimeter of its outer rectangle, and the height-to-width ratio of the target area is greater than 3.5, the construction worker target can be extracted. In the second step, the process of feature extraction of the binary area within the area of ​​the circumscribed rectangle of the personnel image includes: 1. Calculate the gradient of the personnel image in the X and Y coordinate directions to obtain the gradient direction value of each pixel position, and the gradient of the pixel point (xy) in the image satisfies: G x (x,y)=H(x+1,y)-H(x-1,y);G y (x,y)=H(x,y+1)-H(x,y-1); in the above formula, G x (x,y),G y (x, y), H(x, y) represents the gradient of the pixel point (x, y) in the X direction, the gradient in the Y direction and the pixel value respectively; the gradient amplitude G(x, y) at the pixel point (x, y) = (G x (x,y) 2 +G y (x,y) 2 ) 1 / 2; Gradient direction θ(x,y) = tan -1 (G y (x,y) / G x (x, y)); 2. After obtaining the gradient size and direction of the pixel point, since the size of the intercepted image is 64*64, each 8*8 pixels constitutes a unit, the image is regarded as composed of 64 units, and the direction space of each unit in the image is evenly distributed. The range of the direction space is set to 0°-360°, 0°-360° represents the gradient direction space, and the gradient direction is divided into 18 intervals, namely 0°-20°, 20°-40°, and 340°-360°; calculate the gradient direction of all pixels in the unit cell, and perform weighted projection on that direction interval according to the calculated gradient direction, that is, the gradient direction of the pixel belongs to which direction interval, then provide a weight in this direction interval; through step 2, the eigenvector in each unit can be calculated. 3. Every 2*2 unit in the image constitutes a block. The same unit can appear in different blocks. The units in the image are combined into different blocks. All unit feature vectors in a block are connected in series to obtain the directional gradient histogram feature. Since each unit has a 3*3 neighborhood, each unit has 9 features, and each block has 4*9=36 features. With a step size of 8 pixels, there are 7 scanning windows in the horizontal direction and 7 scanning windows in the vertical direction. For a 64*64 image, there are 36*7*7=1764 features. Through the above process, other interference information in the image except the target features is filtered out, which helps to improve the accuracy of personnel recognition. Finally, after extracting the personnel image features, the feature data is trained through the SVM classifier, classified and identified, and the target moving personnel are identified.

[0054] During the safety monitoring process in step 3, a, first identify the helmet and safety belt of the identified personnel, and use the LabelImg labeling tool to label the location and name of the target's helmet and safety belt according to the identified moving target. Then, the detection data of the real bounding boxes annotated in multiple images are obtained as the original data for clustering, and a text file containing the location and category information of the target's real bounding box is generated; each line in the text records the coordinates (x t ,y t ,w t ,h t ),t∈{1,2,...,N}, where (x t ,y t ) is the center point of the bounding box, (w t ,h t ) is the width and height of the border, N is the number of real borders of all targets; b, randomly generate k cluster center point coordinates (W i ,H i), i∈{1,2,...,k}, where W i ,H i is the width and height of the target prior box; c, calculate the distance between the real box of the helmet and the safety belt in each image data set and the adjacent cluster center point, and assign the real box of the helmet and the safety belt target to the adjacent cluster center; d, after all the real box assignments of the targets are completed, recalculate the cluster center for each image data set according to the height and width of the real box of the target, so as to obtain the average value of all the real boxes and the average width W1 i =∑w i / N i , average height H1 i =∑h i / N i ; N i is the target true bounding box of the i-th data set. Repeat steps c and d until the cluster center changes very little and reaches the bottom layer of the cluster. Generate the cluster center coordinates and get the width and height of the helmet and safety belt.

[0055] The process of judging whether a person is wearing a helmet and a safety belt is as follows: First, the OpenPose model is used to detect the human body posture skeleton, and the key node position coordinates and nose coordinates (x 0 ,y 0 ), the left ear coordinate is (x 17 ,y 17 ), right ear(x 16 ,y 16 ), the coordinates of the center point in the calculation are C, (x c ,y c )satisfy:

[0056] (x c ,y c )=(x 0 ,y 0 )+(x 17 ,y 17 )+(x 16 ,y 16 ) / 3; Based on the width and height of the helmet, the center point of the rectangular frame of the helmet is D, (x A ,y A )、(x B ,y B ) are the coordinates of the two points on the diagonal of the rectangular frame of the helmet, then the coordinates of point D (x D ,y D )=(x A ,y A )+(x B ,y B ) / 2; Connect the facial joint point C with the center point D of the helmet to obtain the line segment CD = ((xD ,y D ) 2 -(x C ,y C ) 2 ) 1 / 2 ; The distance between the diagonal lines AB of the rectangular frame of the helmet satisfies AB=((x B ,y B ) 2 -(x A ,y A ) 2 ) 1 / 2 ; Compare the line segment CD with the diagonal AB of the safety helmet rectangular frame. If AB≥CD, it is determined that the worker is wearing a safety helmet. If AB<CD, it is determined that the worker is not wearing a safety helmet. Similarly, obtain the rectangular frame of the safety belt. Through the coordinates of the neck, left shoulder, and right shoulder points, the coordinates of the key nodes of the shoulder can be obtained. By comparing the connection line between the coordinates of the key nodes of the shoulder and the coordinates of the center point of the safety belt with the diagonal distance of the safety belt frame, it can be determined whether the worker is wearing a safety belt. The above process improves the detection accuracy of whether the safety helmet and safety belt are worn. After identifying the workers who are not wearing safety helmets and safety belts, the buzzer of the early warning module will sound an alarm reminder.

[0057] This system reduces the cost of production safety accidents and saves direct economic losses. It improves construction efficiency and shortens the construction period. It improves the safety management level of enterprises and reduces the cost of safety investment. It stimulates the development of the safety monitoring industry and drives related industries; it improves the level of production safety at construction sites and protects the rights and interests of workers. It reduces the occurrence of safety accidents and reduces social risks. It improves the public's awareness of production safety. It improves the safety supervision system and improves supervision capabilities; it reduces accident risks and environmental pollution risks through safety monitoring. New technologies promote resource conservation and environmentally friendly production. The construction of a safe production system promotes green construction.

[0058] Other technical solutions not described in detail in the present invention are all existing technologies in the field and will not be described in detail here.

[0059] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and modifications. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A system for dynamically monitoring the safety of laborers at construction sites, characterized in that: The monitoring platform includes a control host, the control host is connected to a collection module, and the connection and transmission mode between the control host and the collection module adopts a radio signal or a wired electrical signal connection; The control host is provided with a display screen, the control host wireless signal is connected to the mobile terminal, and the control host is also provided with an alarm module, a storage module, and a data analysis module; The collection module is used to collect on-site parameter information of construction personnel; The data analysis module is used to analyze and process the parameters collected by the acquisition module; obtain the construction personnel targets, identify the targets, extract safety features, and judge the safety status of the construction personnel based on the safety features.

2. The system for dynamically monitoring the safety of laborers on construction sites according to claim 1 is characterized in that: The acquisition module includes a plurality of cameras, each of which is electrically connected to a video optical terminal, which is electrically connected to an encoder, and the encoder is connected to a control host of the monitoring platform via a wireless or wired network.

3. The system for dynamically monitoring the safety of laborers on construction sites according to claim 1 is characterized in that: The acquisition module also includes a body temperature measurement sensor and a heart rate and blood oxygen sensor. The body temperature measurement sensor and heart rate and blood oxygen sensor acquisition module is integrated with an A / D converter, which is converted into a digital signal through the A / D converter. The acquisition module transmits the digital signal to the control host through the integrated LoRa wireless communication mode, and the control host receives the physiological signal. The acquisition module is also integrated with a GPS positioning function.

4. The system for dynamically monitoring the safety of laborers on construction sites according to claim 3 is characterized in that: The LoRa wireless communication module is the ATK-LORA-01 wireless serial port module, which is powered by 5V. The module operates at a frequency of 410-441MHz, in 1MHz increments, with a total of 32 channels. It has multiple working modes and uses an SMA antenna with a maximum transmission distance of 3km. The module's TXD port is connected to the PB11 port of the main control chip, the RXD port is connected to the PB10 port of the main control chip, the AUX port is connected to the PA4 port of the main control chip, and the MDO port is connected to the PA15 port of the main control chip. Directional transmission is used to set the target node address and channel to transmit the collected vital sign data.

5. The system for dynamically monitoring the safety of laborers at a construction site according to claim 1 is characterized in that: The alarm module of the control host includes a buzzer, which receives instructions from the control host; the storage module stores the collected information in the control host.

6. A readable storage medium for dynamically monitoring the safety of laborers at a construction site, characterized in that: Control instructions are stored in the readable storage medium, and the readable storage medium is used in the system for dynamically monitoring the safety of laborers on a construction site as described in any one of claims 1-5.

7. A method for dynamically monitoring the safety of laborers at a construction site, using a system for dynamically monitoring the safety of laborers at a construction site as claimed in any one of claims 1 to 5, characterized in that: The method includes: Step 1: Information collection, which includes image collection and physiological characteristics collection. Image collection is carried out by using a camera to collect images of the construction site. Physiological characteristics collection is carried out by using a temperature detection sensor and a heart rate and blood oxygen sensor to collect the temperature, heart rate and blood oxygen concentration of the laborers. The temperature detection sensor is set inside the laborers' helmets to collect the temperature of the laborers, and the heart rate and blood oxygen sensor is set inside the laborers' safety belts near the heart. Step 2: The collected temperature data and heart rate oxygen data of the laborers are transmitted to the control host, which processes the collected data and makes a judgment. When the set temperature and heart rate oxygen data thresholds are reached, the control host sends a command to the alarm module to sound an alarm. Step 3: After obtaining the construction site image through the camera, the control center pre-processes the image, performs moving target detection, target recognition, and safety detection. If the detection data result shows that the laborers do not wear safety belts and helmets properly, a command is sent to the alarm module and the buzzer issues an early warning.

8. The method for dynamically monitoring the safety of laborers at a construction site according to claim 7, characterized in that: During image information collection, after the control center obtains the on-site image information, it first determines whether a danger zone has been set for the obtained image. If a danger zone has been set for the work site, the control center saves a template image with the danger zone set. After obtaining the image, it compares the obtained image information with the saved template image of the danger zone, and performs moving target recognition and safety detection on the labor personnel in the danger zone. If no danger zone has been set for the work site, moving target recognition and safety detection are performed on the labor personnel in the entire image.

9. The method for dynamically monitoring the safety of laborers at a construction site according to claim 8, characterized in that: The setting of the danger zone is to obtain information of a picture, and in the process of processing the picture, obtain data of four boundary fixed points of the danger zone of the image, obtain the coordinates of the four points, and connect the four coordinate points into a quadrilateral area, which is the danger zone.

10. The method for dynamically monitoring the safety of laborers at a construction site according to claim 7, characterized in that: In step 2, the temperature, heart rate and blood oxygen sensor data are collected every 200ms, and the control center obtains the location information of the labor personnel through the GPS function. The GPS collects the location information every 5s and transmits it to the control center; the control center saves the sensor data collected in the last two time intervals in the corresponding array, and removes abnormal data through data preprocessing; after the data collection of this time interval is completed, the serial port interrupt is disabled, the reception of sensor data is suspended, and the sensor data in this time interval is obtained as the final transmission data, and the data is displayed on the display screen and the mobile terminal.

Images