Intelligent construction site remote monitoring system based on Internet of Things

Abstract

The invention relates to the Internet of Things video analysis technology, in particular to an intelligent construction site remote monitoring system based on the Internet of Things, which is characterized in that firstly, a risk point location analysis module is combined with historical fault data to intelligently analyze the risk index of each region, and then a data acquisition module is used for collecting personnel information and video monitoring data; the video analysis module analyzes the video monitoring data and records the troubleshooting time of the staff in each region, and the evaluation accuracy analysis module adjusts the corresponding standard troubleshooting time in combination with the regional risk index. And the evaluation accuracy of the working personnel on the target area is calculated by considering the influence of the working experience of the working personnel on the troubleshooting accuracy, so that a more objective and accurate evaluation result is obtained. The warning module can give an alarm in time when it is found that the evaluation accuracy is insufficient according to a set accuracy threshold value, so that correction measures are taken in time when the evaluation accuracy of workers is insufficient, and the safety management level of the construction site is greatly improved.

Description

Technical Field

[0001] The present invention relates to the Internet of Things video analysis technology, and specifically to an intelligent construction site remote monitoring system based on the Internet of Things. Background Art

[0002] In the daily operation of a construction site, risk management and the investigation of potential safety hazards are important links to ensure the safety of workers and the smooth progress of the project. The existing investigation of potential risks and hazards at the construction site mainly relies on manual inspections. This mode requires safety supervisors to conduct on-site inspections of the construction site regularly or irregularly to identify potential risk points and safety hazards. However, manual risk identification often requires staff to have sufficient work experience, and human negligence is difficult to avoid. For example, staff often miss some risk points due to fatigue, inattention, etc., making the risk assessment of the target area inaccurate, which will cause great loopholes in the construction safety of the construction site. Therefore, there is an urgent need for a remote monitoring system that can evaluate the risk investigation work of staff. Summary of the Invention

[0003] The technical problem solved by the present invention is to provide an intelligent construction site remote monitoring system based on the Internet of Things, which solves the problem of inaccurate evaluation caused by human negligence due to fatigue, inattention, etc. in the manual risk assessment of the target area.

[0004] The basic solution provided by the present invention: An intelligent construction site remote monitoring system based on the Internet of Things, including a data collection module, a risk point analysis module, a video analysis module, an evaluation accuracy analysis module, and a warning module;

[0005] The data collection module includes a personnel information collection module and a video monitoring module. The personnel information collection module is used to collect staff information, and the video monitoring module is used to collect the monitoring video data of the working area;

[0006] The risk point analysis module is used to analyze the accident risk index R of each area according to historical failure data:

[0007]

[0008] In the formula, n is the number of historical failure types in the target area, i is the ordinal number in the historical failure types, λ i is the frequency of a certain type of failure occurring in the target area, and S i is the matching factor of the impact caused by the occurrence of this type of failure;

[0009] The video analysis module is used to analyze the investigation time t of the staff in each area based on the monitoring video data. The evaluation accuracy analysis module is used to analyze the evaluation accuracy A of the staff for the target area according to the staff information, the regional risk index, and the corresponding investigation time:

[0010]

[0011] In the formula, t0 is the standard investigation time, t is the actual investigation time, T is the user's working years, T max is the maximum working years, k is the growth rate of the investigation proficiency, P is the historical evaluation accuracy of the staff, R est is the accident risk level of the target area, and the accident risk level R est is divided according to the accident risk index R;

[0012] The warning module is provided with an accuracy threshold k. When the evaluation accuracy A does not reach the evaluation accuracy threshold k, the warning module issues a warning that the investigation does not meet the standard.

[0013] Furthermore, the video analysis module includes a preprocessing module, an identification module, and an investigation record module;

[0014] The preprocessing module is used to perform preprocessing operations on the monitoring video data to improve the image quality; 1

[0015] The identification module is used to identify the personnel identity information and risk investigation behaviors according to the preprocessed video images. When the target personnel are identified as non-staff, the warning module issues a warning message for non-staff;

[0016] The investigation record module is used to record the area where the personnel are located and the investigation time t that the risk investigation behavior in this area continues.

[0017] Furthermore, the identification module includes an identity identification module and a behavior identification module;

[0018] The identity identification module is used to identify the personnel identity information according to the preprocessed video images;

[0019] The behavior identification module is used to identify the risk investigation behaviors of the target personnel through the risk investigation behavior model. The establishment process of the risk investigation behavior model includes the steps of:

[0020] Extracting the risk investigation behavior characteristics from the preprocessed video images;

[0021] Combining the extracted risk investigation behavior characteristics into an action sequence, and each sequence represents a complete behavior process;

[0022] Using a recurrent neural network to train and identify the action sequence to obtain a risk investigation behavior model.

[0023] Furthermore, the data acquisition module further includes an environmental data acquisition module, which includes a number of temperature sensors, humidity sensors, smoke sensors and flame sensors distributed in various areas of the construction site; the environmental data acquisition module collects temperature data, humidity data and smoke concentration data of each area through the temperature sensors, humidity sensors and smoke sensors, and detects whether there is open fire in the target area through the flame sensor; when open fire is detected in the target area, the warning module issues a fire alarm for prompt.

[0024] Furthermore, it further includes an information display module, which includes a video display module, an information marking module and a prompt module;

[0025] The video display module is used to display the monitoring videos of each area; the information marking module is used to mark the collected environmental data on the monitoring video corresponding to the target area; the prompt module is used to flash and prompt the monitoring videos corresponding to the areas where the inspection fails to meet the standard and the areas where open fire appears.

[0026] Furthermore, it further includes an inspection record module, which includes a positioning module and an inspection confirmation module. The positioning module is used to obtain the positioning information of the staff, and the inspection confirmation module is used to determine the working area where the staff is located according to the staff positioning information, obtain the historical fault types of the area according to the working area and generate a corresponding risk inspection record form, and confirm the risk inspection of the current area by recording the inspection results and time in the risk record project form.

[0027] Furthermore, the inspection record module further includes a recheck module, which is used to send a non-compliance prompt to the staff in the corresponding area of the risk record project form according to the non-compliance warning issued by the warning module, reminding the staff to conduct a risk inspection and confirmation again.

[0028] The principle and advantages of the present invention are as follows: Firstly, this solution uses the risk point analysis module to intelligently analyze the risk index of each area in combination with historical fault data, and then collects personnel information and video monitoring data through the data acquisition module, analyzes the video monitoring data through the video analysis module and records the inspection time of the staff in each area. The evaluation accuracy analysis module adjusts the corresponding standard inspection time in combination with the regional risk index, and then calculates the evaluation accuracy of the staff for the target area by considering the influence of the staff's work experience on the inspection accuracy, so as to obtain a more objective and accurate evaluation result. The warning module can issue a warning in time when the inspection finds that the evaluation accuracy is insufficient according to the set accuracy threshold, so as to take corrective measures in time when the evaluation accuracy of the staff is insufficient, greatly improving the level of construction site safety management and ensuring the safety of workers and the smooth progress of the project. Brief Description of the Drawings

[0029] Figure 1 This is a logic block diagram of an embodiment of an Internet of Things-based intelligent construction site remote monitoring system of the present invention. Specific implementation mode

[0030] The following is a further detailed description through specific implementation modes:

[0031] The specific implementation process is as follows:

[0032] Embodiment 1

[0033] As shown in the appendix of Embodiment 1 Figure 1 An Internet of Things-based intelligent construction site remote monitoring system includes a data collection module, a risk point analysis module, a video analysis module, an evaluation accuracy analysis module, and a warning module. The system automatically collects staff information and monitoring video data through the data collection module, and uses the risk point analysis module to analyze historical fault data to evaluate the accident risk index of each area. At the same time, the video analysis module can automatically record the inspection time of the staff in each area, and the evaluation accuracy analysis module analyzes the evaluation accuracy of the staff for the target area by combining the staff information, area risk index, and inspection time. The warning module determines whether to issue a warning for unqualified inspections according to the set accuracy threshold. Such a system not only improves the real-time performance and accuracy of monitoring, but also can take corrective measures in time when the evaluation accuracy of the staff is insufficient, improving the efficiency and effect of safety management.

[0034] Specifically, the data collection module includes a personnel information collection module and a video monitoring module. The personnel information collection module is used to collect staff information including working years. The video monitoring module uses high-definition cameras to monitor the working area in real time, ensuring that all key equipment and construction areas are covered, and collecting monitoring video data of the working area through high-definition cameras.

[0035] The risk point analysis module is used to analyze the accident risk index R of each area according to historical fault data:

[0036]

[0037] In the formula, n is the number of historical fault types in the target area, i is the ordinal number in the historical fault type, and λ i is the frequency (per year) of a certain type of fault occurring in the target area, and S i is the matching factor of the impact caused by this type of fault. The accident risk index of each area (such as earth excavation area, formwork production area, tower crane, etc.) is analyzed through the system. The comparison table of faults and impact matching factors is as follows:

[0038] Table 1 Comparison table of faults and impact matching factors

[0039] Accident consequences Impact matching factor Minor equipment damage 0.3 Moderate equipment damage 1 Severe equipment damage 2 Project delay (minor delay) 0.3 Project delay (medium delay) 1 Project delay (major delay) 3 Minor injury to personnel 1 Serious injury to personnel 2 Fatal injury to personnel 3

[0040] The video analysis module is used to analyze the inspection time t of the staff in each area according to the surveillance video data. Specifically, the video analysis module includes a preprocessing module, an identification module, and an inspection record module.

[0041] The preprocessing module is used to perform preprocessing operations on the surveillance video data. In this embodiment, the surveillance video image is first subjected to preprocessing operations such as noise reduction and contrast enhancement to improve the image quality. Then, image segmentation technology is applied to separate different objects and the background in the image for subsequent feature extraction.

[0042] The identification module includes an identity identification module and a behavior identification module. The identity identification module is used to identify the identity information of personnel according to the preprocessed video image. When a staff member enters the working area, a real-time image of the staff member is captured by a high-definition camera. When the target person is identified as a non-staff member, the warning module issues a non-staff member warning message.

[0043] The behavior identification module is used to input the preprocessed video image into the risk inspection behavior model to identify the risk inspection behavior of the target person. The establishment process of the risk inspection behavior model includes the steps:

[0044] First, extract the risk inspection behavior features from the preprocessed video image. Determine the key features that can represent the risk inspection behavior, such as gestures, postures, and tools used, etc. These features should be able to distinguish different risk inspection behaviors and be consistent under different environments and conditions. After determining the features, apply computer vision techniques, such as edge detection, contour extraction, and image segmentation, to identify and extract the selected features.

[0045] Secondly, segment the extracted risk inspection behavior features into independent action units according to consecutive video frames. Each unit represents a basic behavior action. Combine these action units into an action sequence in chronological order, and label the action sequence to indicate the start and end points of each sequence, as well as the key steps in the middle. Each sequence represents a complete behavior process.

[0046] Finally, use a recurrent neural network to train and identify the action sequence. Train the RNN through the labeled action sequence dataset, adjust the network parameters to minimize the error, and then adopt the backpropagation algorithm and gradient descent optimization method to continuously update the network weights until the model converges to obtain the risk inspection behavior model.

[0047] The inspection record module is used to record the area where the person is located and the inspection time t during which the risk inspection behavior in that area continues after identifying the risk inspection behavior of the person.

[0048] The evaluation accuracy analysis module is used to analyze the evaluation accuracy A of the staff for the target area according to the staff information, the regional risk index, and the corresponding investigation time:

[0049]

[0050] In the formula, t0 is the standard investigation time, t is the actual investigation time, T is the user's working years, and T max is the maximum working years, k is the growth rate of the investigation proficiency, P is the historical evaluation accuracy of the staff. In this embodiment, the minimum working years of the user is five years. The user's working years and the historical evaluation accuracy of the staff are obtained from the enterprise database. Since the longer the user's working years, the richer the experience, and the stronger the judgment of environmental hazards, this solution substitutes the user's own experience and its historical evaluation accuracy to judge the accuracy of its current evaluation, which can make the judgment result more real and accurate. In addition, in this embodiment, the growth rate k of the investigation proficiency is obtained according to the staff's working years, investigation speed, and evaluation accuracy in the historical records. R est is the accident risk level of the target area, and the accident risk level R est is divided according to the accident risk index R. When R = 0, the accident risk level is risk-free, and R est = 1; when R < 1, the accident risk level is low risk, and R est = 1.25; when 1 ≤ R < 2, the accident risk level is medium risk, and R est = 1.5; when R > 2, the accident risk level is high risk, and R est = 2. Controlling the standard investigation time in combination with the actual risk level of the target area can make the analysis result more in line with the actual situation and make the judgment result more accurate.

[0051] The warning module is provided with an accuracy threshold k. When the evaluation accuracy A does not reach the evaluation accuracy threshold k, the warning module issues a warning that the investigation is not up to standard, enabling the staff to take corrective measures in time when the evaluation accuracy is insufficient, greatly improving the level of construction site safety management and ensuring the safety of workers and the smooth progress of the project..

[0052] In addition, this embodiment further includes an environmental data acquisition module and an information display module. The environmental data acquisition module includes a number of temperature sensors, humidity sensors, smoke sensors, and flame sensors distributed in various areas of the construction site. The environmental data acquisition module collects temperature data, humidity data, and smoke concentration data of each area through the temperature sensors, humidity sensors, and smoke sensors, and detects whether there is open fire in the target area through the flame sensor. When an open fire is detected in the target area, the data acquisition module sends a fire alarm signal to the warning module, and the warning module issues a fire alarm for prompt after receiving the fire alarm signal. As the construction project progresses, by continuously arranging temperature sensors, humidity sensors, smoke sensors, and flame sensors in key areas of the construction site, the system can monitor the environmental conditions of the construction site in real time, including temperature, humidity, smoke concentration, and open fire situation. In addition, in this embodiment, there is also a smart safety helmet equipped with sensors and a video monitoring module for mobile environmental data acquisition and monitoring. This comprehensive monitoring helps to promptly detect potential environmental risks, such as fire hazards, and other environmental factors that may affect the health and safety of workers.

[0053] The information display module includes a video display module, an information marking module, and a prompt module. The video display module is used to display the monitoring videos of each area. The information marking module is used to mark the collected environmental data on the monitoring video corresponding to the target area. The prompt module is used to flash and prompt the monitoring videos corresponding to the areas where non-compliance is detected and open fire appears. The information display module in this embodiment provides an intuitive and easy-to-understand user interface through the video display module, the information marking module, and the prompt module. The environmental data (such as temperature, humidity, smoke concentration) marked on the monitoring video and the flashing prompt of non-compliant or dangerous areas not only help the staff quickly locate the problems, but also enhance the visual impact of safety issues, thereby improving the safety awareness of construction site personnel and the efficiency of risk management. This instant feedback mechanism enables safety supervisors and construction site managers to quickly identify problem areas, make corresponding decisions and take actions, encourages the implementation of timely corrective and preventive measures, and reduces the occurrence of safety accidents.

[0054] Embodiment Two

[0055] The difference between the second embodiment and the first embodiment is only that the second embodiment further includes a troubleshooting record module, which is implemented through an APP installed on a mobile phone. The troubleshooting record module includes a positioning module, a troubleshooting confirmation module, and a prompt module. The positioning module is used to obtain the positioning information of the staff. The troubleshooting confirmation module is used to determine the working area where the staff is located according to the positioning information of the staff, obtain the historical fault types in this area according to the working area, and generate a corresponding risk troubleshooting record form, and confirm the risk troubleshooting of the current area by recording the troubleshooting results and time in the risk record item form. The review module is used to send a non-compliance prompt to the staff in the corresponding area of the risk record item form according to the non-compliance warning issued by the warning module, reminding the staff to conduct a risk troubleshooting confirmation again to avoid potential risks missed due to incomplete initial troubleshooting.

[0056] The above are only embodiments of the present invention. Specific structures and common knowledge such as characteristics well known in the art are not described in detail here. Those of ordinary skill in the art know all the common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can know all the existing technologies in this field, and have the ability to apply conventional experimental means before this date. Those of ordinary skill in the art can, under the inspiration given in this application, combine their own abilities to improve and implement this solution. Some typical well-known structures or well-known methods should not become obstacles for those of ordinary skill in the art to implement this application. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can still be made, which should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners described in the specification can be used to explain the content of the claims.

Claims

1. An intelligent construction site remote monitoring system based on the Internet of Things, characterized in that: It includes a data collection module, a risk point analysis module, a video analysis module, an evaluation accuracy analysis module, and a warning module; The data collection module includes a personnel information collection module and a video monitoring module. The personnel information collection module is used to collect staff information, and the video monitoring module is used to collect monitoring video data of the working area; The risk point analysis module is used to analyze the accident risk index R of each area according to historical failure data: where n is the number of historical fault types in the target area, i is the ordinal number in the historical fault types, and λ i is the frequency of a certain type of fault occurring in the target area, and S i is the matching factor of the impact caused by this type of fault; The video analysis module is used to analyze the inspection time t of the staff in each area based on the monitoring video data. The evaluation accuracy analysis module is used to analyze the evaluation accuracy A of the staff for the target area according to the staff information, the area risk index, and the corresponding inspection time: Where, \(t_0\) is the standard investigation time, \(t\) is the actual investigation time, \(T\) is the user's working years, and \(T\) max is the maximum working years, \(k\) is the growth rate of investigation proficiency, \(P\) is the historical evaluation accuracy of the staff, and \(R\) est is the accident risk level of the target area. The accident risk level \(R\) est is divided according to the accident risk index \(R\); The warning module is provided with an accuracy threshold k. When the evaluation accuracy A does not reach the evaluation accuracy threshold k, the warning module issues a warning that the inspection is not up to standard.

2. The intelligent construction site remote monitoring system based on the Internet of Things according to claim 1, characterized in that: The video analysis module includes a preprocessing module, an identification module, and an inspection record module; The preprocessing module is used to perform preprocessing operations on the monitoring video data to improve the image quality; The identification module is used to identify the personnel identity information and risk inspection behavior according to the preprocessed video image. When the target person is identified as a non-staff member, the warning module issues a non-staff warning message; The inspection record module is used to record the area where the personnel are located and the inspection time t during which the risk inspection behavior in that area continues; 3. The intelligent construction site remote monitoring system based on the Internet of Things according to claim 2, characterized in that: The identification module includes an identity identification module and a behavior identification module; The identity identification module is used to identify the personnel identity information according to the preprocessed video image; The behavior identification module is used to identify the risk inspection behavior of the target person through a risk inspection behavior model. The establishment process of the risk inspection behavior model includes the steps of: Extracting risk inspection behavior features from the preprocessed video image; Combining the extracted risk inspection behavior features into action sequences, and each sequence represents a complete behavior process; Using a recurrent neural network to train and identify the action sequences to obtain a risk inspection behavior model.

4. The intelligent construction site remote monitoring system based on the Internet of Things according to claim 3, wherein: The data collection module further includes an environmental data collection module. The environmental data collection module includes a number of temperature sensors, humidity sensors, smoke sensors, and flame sensors distributed in each area of the construction site. The environmental data collection module collects temperature data, humidity data, and smoke concentration data of each area through the temperature sensors, humidity sensors, and smoke sensors, and detects whether there is an open fire in the target area through the flame sensor. When an open fire is detected in the target area, the warning module issues a fire alarm for prompt.

5. The intelligent construction site remote monitoring system based on the Internet of Things according to claim 4, characterized in that: It further includes an information display module. The information display module includes a video display module, an information marking module, and a prompt module; The video display module is used to display the monitoring videos of each area. The information marking module is used to mark the collected environmental data on the monitoring video corresponding to the target area; The prompt module is used to flash the monitoring videos corresponding to the areas where the inspection is not up to standard and the areas where there is an open fire.

6. The intelligent construction site remote monitoring system based on the Internet of Things according to claim 1, characterized in that: It further includes an investigation record module, which includes a positioning module and an investigation confirmation module. The positioning module is used to obtain the positioning information of the staff. The investigation confirmation module is used to determine the working area where the staff is located according to the staff positioning information, obtain the historical fault types in this area based on the working area, and generate a corresponding risk investigation record form, and confirm the risk investigation of the current area by recording the investigation results and time in the risk record item form.

7. The intelligent construction site remote monitoring system based on the Internet of Things according to claim 6, wherein: The investigation record module further includes a recheck module, which is used to send a non-compliance prompt to the staff in the corresponding area of the risk record item form according to the non-compliance warning issued by the warning module, reminding the staff to conduct risk investigation and confirmation again.

Images