Construction personnel state monitoring and early warning system and method based on attitude tracking

Abstract

The invention discloses a construction personnel state monitoring and early warning system and method based on attitude tracking, and relates to the technical field of monitoring and early warning, the system comprises a data acquisition module, an attitude tracking module, a behavior analysis module and an alarm module which are connected in sequence; the data acquisition module is used for capturing a real-time video stream of a constructor; the attitude tracking module is used for extracting skeleton key points of the constructor from each frame of image in the real-time video stream by using a computer vision algorithm to obtain coordinates of the skeleton key points of the constructor; a posture recognition model is deployed in the behavior analysis module, coordinates of the skeleton key points are input into the posture recognition model, actions of construction personnel are determined, unsafe behavior detection is carried out, and an unsafe behavior recognition result is obtained; and the alarm module is used for triggering an alarm when the unsafe behavior identification result is an unsafe behavior. According to the invention, the real-time performance, precision and automation degree of monitoring and safety alarm of constructors are improved.

Description

Technical Field

[0001] The present application relates to the field of monitoring and early warning technology, and in particular to a construction personnel status monitoring and early warning system and method based on posture tracking. Background Art

[0002] At present, construction site safety management mainly relies on the following technical means: manual inspection and supervision, video surveillance, sensor-based monitoring, and computer vision and artificial intelligence.

[0003] However, with the development of the construction industry, the complexity of construction sites and the increase in personnel mobility, traditional safety management methods have been unable to meet the needs of efficient and safe management. Existing technical solutions have the following problems in construction site safety management: insufficient real-time performance; low accuracy; low degree of automation and poor scalability. Summary of the invention

[0004] The purpose of this application is to provide a construction personnel status monitoring and early warning system and method based on posture tracking, so as to solve the problems of insufficient real-time performance, low accuracy and low degree of automation in existing construction site safety management methods.

[0005] To achieve the above objectives, this application provides the following solutions.

[0006] In a first aspect, the present application provides a construction worker status monitoring and early warning system based on posture tracking, comprising: a data acquisition module, a posture tracking module, a behavior analysis module and an alarm module connected in sequence; The data acquisition module is used to capture the real-time video stream of the construction personnel and send the real-time video stream to the posture tracking module; The posture tracking module is used to extract the skeleton key points of the construction personnel from each frame image in the real-time video stream by using a computer vision algorithm, obtain the coordinates of the skeleton key points of the construction personnel, and send the coordinates of the skeleton key points to the behavior analysis module; A posture recognition model is deployed in the behavior analysis module, and the behavior analysis module is used to input the coordinates of the skeleton key points into the posture recognition model, determine the actions of the construction personnel, and use the actions of the construction personnel to detect unsafe behaviors, obtain unsafe behavior recognition results, and send the unsafe behavior recognition results to the alarm module; the posture recognition model is obtained by training a machine learning model using a training data set, and the training data set includes the coordinates of multiple skeleton key points and corresponding actions; the unsafe behavior recognition result is an unsafe behavior or a safe behavior; The alarm module is used to trigger an alarm when the unsafe behavior identification result is an unsafe behavior.

[0007] Optionally, the construction personnel status monitoring and early warning system based on posture tracking further includes: a data storage and analysis module; the data acquisition module, the posture tracking module, the behavior analysis module and the alarm module are respectively connected to the data storage and analysis module; The data storage and analysis module is used to store the real-time video stream, the coordinates of the skeleton key points, the unsafe behavior recognition results and the alarm log in the database, and regularly analyze the data stored in the database to generate a safety management report.

[0008] Optionally, the computer vision algorithm is OpenPose technology.

[0009] Optionally, the OpenPose technology is used to extract the skeleton key points of the construction worker from any image to obtain the coordinates of the skeleton key points of the construction worker, including: Preprocessing the image to obtain a preprocessed image; the preprocessing includes: adjusting the image size and color channel conversion; The preprocessed image is input into the feature extraction network to obtain multiple feature maps; the number of feature maps is equal to the number of skeleton key points; For each feature map, obtain the heat map corresponding to the key points of the skeleton; Based on the pixel values ​​at all pixels in the heat map of each bone key point, the coordinates of the corresponding bone key point are determined.

[0010] Optionally, unsafe behavior detection is performed using the actions of the construction workers to obtain unsafe behavior identification results, including: The actions of the construction workers are compared with the unsafe actions in the predefined unsafe behavior library to obtain the unsafe behavior identification result.

[0011] Optionally, the actions of the construction workers are compared with unsafe actions in a predefined unsafe behavior library to obtain the unsafe behavior identification result, including: The similarity between the construction worker's action and the unsafe action is calculated based on the feature vector of the construction worker's action and the feature vector of the unsafe action using the cosine similarity calculation formula; the cosine similarity calculation formula is: ; in, is the similarity between the construction worker's actions and unsafe actions; is the characteristic vector of the construction worker’s action; is the feature vector of the unsafe action; is the dot product; is the magnitude of the vector; If the similarity is greater than a preset similarity threshold, the unsafe behavior recognition result is determined as an unsafe behavior; Otherwise, the unsafe behavior identification result is determined as a safe behavior.

[0012] Optionally, trigger an alarm, including: Alert safety supervisors via voice, text message or mobile app.

[0013] In a second aspect, the present application provides a construction worker status monitoring and early warning method based on posture tracking, and a construction worker status monitoring and early warning system based on posture tracking is implemented. The construction worker status monitoring and early warning method based on posture tracking includes: Capture live video streams of construction crews; Using a computer vision algorithm, extracting skeleton key points of the construction worker from each frame image in the real-time video stream to obtain the coordinates of the skeleton key points of the construction worker; The coordinates of the skeleton key points are input into the posture recognition model to determine the actions of the construction workers, and the actions of the construction workers are used to detect unsafe behaviors to obtain unsafe behavior recognition results; the posture recognition model is obtained by training a machine learning model using a training data set, and the training data set includes the coordinates of multiple skeleton key points and corresponding actions; the unsafe behavior recognition result is an unsafe behavior or a safe behavior; When the unsafe behavior identification result is an unsafe behavior, an alarm is triggered.

[0014] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-mentioned methods for monitoring and early warning of the status of construction personnel based on posture tracking.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-mentioned methods for monitoring and early warning of the status of construction personnel based on posture tracking.

[0016] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned methods for monitoring and early warning of the status of construction personnel based on posture tracking.

[0017] According to the specific embodiments provided in this application, this application discloses the following technical effects: The present application discloses a construction worker status monitoring and early warning system and method based on posture tracking, the system comprises: a data acquisition module, a posture tracking module, a behavior analysis module and an alarm module connected in sequence; the data acquisition module is used to capture the real-time video stream of the construction workers; the posture tracking module is used to extract the skeleton key points of the construction workers from each frame image in the real-time video stream by using a computer vision algorithm, and obtain the coordinates of the skeleton key points of the construction workers; a posture recognition model is deployed in the behavior analysis module, and the behavior analysis module is used to input the coordinates of the skeleton key points into the posture recognition model, determine the actions of the construction workers, and use the actions of the construction workers to detect unsafe behaviors and obtain unsafe behavior recognition results; the posture recognition model is obtained by training a machine learning model using a training data set, and the training data set includes the coordinates of multiple skeleton key points and corresponding actions; the unsafe behavior recognition result is an unsafe behavior or a safe behavior; the alarm module is used to trigger an alarm when the unsafe behavior recognition result is an unsafe behavior. This application improves the real-time performance of construction worker monitoring and safety alarms by integrating a data acquisition module and a posture tracking module; utilizes computer vision algorithms and posture recognition models to determine the movements of construction workers and thereby identify unsafe behaviors, thereby improving the accuracy of construction worker monitoring and safety alarms; and improves the degree of automation of construction worker monitoring and safety alarms through the coordination of various modules. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A schematic diagram of the structure of a construction personnel status monitoring and early warning system based on posture tracking provided in one embodiment of the present application.

[0020] Figure 2 A schematic diagram of the structure of a computer device provided in one embodiment of the present application.

[0021] Reference numerals: Data acquisition module—1, posture tracking module—2, behavior analysis module—3, alarm module—4, data storage and analysis module—5. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] The purpose of this application is to provide a construction personnel status monitoring and early warning system and method based on posture tracking, aiming to improve the real-time, accuracy and automation of construction personnel monitoring and safety alarms.

[0024] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0025] In an exemplary embodiment, Figure 1 As shown, the construction personnel status monitoring and early warning system based on posture tracking in this embodiment includes: a data acquisition module 1, a posture tracking module 2, a behavior analysis module 3 and an alarm module 4 connected in sequence.

[0026] The data acquisition module 1 is used to capture the real-time video stream of the construction workers and send the real-time video stream to the posture tracking module 2.

[0027] Specifically, the data acquisition module 1 includes but is not limited to: a camera and a sensor.

[0028] The posture tracking module 2 is used to extract the skeleton key points of the construction workers from each frame image in the real-time video stream using a computer vision algorithm, obtain the coordinates of the skeleton key points of the construction workers, and send the coordinates of the skeleton key points to the behavior analysis module 3.

[0029] A posture recognition model is deployed in the behavior analysis module 3. The behavior analysis module 3 is used to input the coordinates of the skeleton key points into the posture recognition model, determine the actions of the construction workers, and use the actions of the construction workers to detect unsafe behaviors, obtain unsafe behavior recognition results, and send the unsafe behavior recognition results to the alarm module 4; the posture recognition model is obtained by training the machine learning model using a training data set, and the training data set includes the coordinates of multiple skeleton key points and corresponding actions; the unsafe behavior recognition result is an unsafe behavior or a safe behavior.

[0030] The alarm module 4 is used to trigger an alarm when the unsafe behavior identification result is an unsafe behavior.

[0031] As an optional implementation, the construction personnel status monitoring and early warning system based on posture tracking also includes: a data storage and analysis module 5; a data acquisition module 1, a posture tracking module 2, a behavior analysis module 3 and an alarm module 4 are respectively connected to the data storage and analysis module 5.

[0032] The data storage and analysis module 5 is used to store the real-time video stream, the coordinates of the skeleton key points, the unsafe behavior recognition results and the alarm log in the database, and regularly analyze the data stored in the database to generate a safety management report.

[0033] As an optional implementation, the computer vision algorithm is OpenPose technology.

[0034] As an optional implementation, using OpenPose technology to extract skeleton key points of construction workers from any image to obtain the coordinates of the skeleton key points of the construction workers includes the following steps.

[0035] Step S1: preprocessing the image to obtain a preprocessed image; the preprocessing includes: adjusting the image size and color channel conversion.

[0036] Specifically, when resizing an image, the image is scaled to a specific size to ensure that it matches the input requirements of the feature extraction network. When converting color channels, the image is converted from BGR format to RGB format.

[0037] Step S2: Input the preprocessed image into the feature extraction network to obtain multiple feature maps; the number of feature maps is equal to the number of skeleton key points.

[0038] Specifically, feature extraction networks include but are not limited to: Convolutional Neural Networks (CNN), VGG networks and ResNet networks.

[0039] Step S3: For each feature map, obtain the heat map corresponding to the skeleton key point.

[0040] Specifically, after a series of upsampling (deconvolution or interpolation), each feature map is output as multiple heat maps, usually one heat map for each skeletal key point.

[0041] .

[0042] in, is the heat map of the kth skeleton key point; For upsampling; It is the feature map corresponding to the skeleton key points.

[0043] Step S4: Based on the pixel values ​​at all pixels in the heat map of each bone key point, determine the coordinates of the corresponding bone key point.

[0044] Specifically, the pixel value at each pixel point of the heat map represents the possibility that the pixel point is a skeleton key point, and the higher the pixel value, the greater the possibility.

[0045] Step S4 specifically includes: Step S41: Find the pixel with the largest pixel value in the heat map of each skeleton key point respectively, and determine the position of the pixel with the largest pixel value as the coordinate of the corresponding skeleton key point. The formula is: .

[0046] in, is the coordinate of the kth bone key point.

[0047] After step S41, the method further includes: Step S42: Connect the bone key points into a bone structure according to their connectivity (such as shoulder to elbow, elbow to wrist, etc.), and use a connection matrix or graph to describe the relationship between the bone key points for subsequent visualization and analysis.

[0048] Step S43: Convert the coordinates of the skeleton key points to the coordinates of the original image (i.e., the image acquired by the data acquisition module) so as to correctly mark the image. By setting a confidence threshold, only the skeleton key points with a confidence higher than the confidence threshold are retained to reduce false detection.

[0049] Apply smoothing algorithms (such as Kalman filter) to process the video stream to reduce the position jitter of the skeleton key points and improve the stability of the detection results.

[0050] Step S44: Return the coordinates of all skeleton key points and mark the detected skeleton key points on the original image. Different colors or sizes are usually used to represent skeleton key points and bones of different confidence levels for subsequent analysis or visualization.

[0051] Step S45: using the coordinate translation method or the velocity and acceleration method, the coordinates of the skeleton key points are converted into feature vectors.

[0052] Among them, the coordinate translation method normalizes the coordinates of the key points of the skeleton or calculates the relative position to remove the influence of body size and position. The velocity and acceleration method calculates the velocity and acceleration of the key points of the skeleton through time series data to enhance the expressiveness of the feature vector.

[0053] As an optional implementation method, unsafe behavior detection is performed using the actions of construction workers to obtain unsafe behavior identification results, including: The actions of the construction workers are compared with the unsafe actions in the predefined unsafe behavior library to obtain the unsafe behavior identification results.

[0054] As an optional implementation, the actions of the construction workers are compared with the unsafe actions in the predefined unsafe behavior library to obtain unsafe behavior identification results, including: The cosine similarity calculation formula is used to calculate the similarity between the construction worker's action and the unsafe action based on the feature vector of the construction worker's action and the feature vector of the unsafe action; the cosine similarity calculation formula is: ; in, is the similarity between the construction worker's actions and unsafe actions; is the characteristic vector of the construction worker’s action; is the feature vector of the unsafe action; is the dot product; is the magnitude of the vector.

[0055] .

[0056] .

[0057] .

[0058] in, for The i-th dimension component of ; for The i-th dimension component of ; is the dimension of the feature vector.

[0059] If the similarity is greater than a preset similarity threshold, the unsafe behavior recognition result is determined as an unsafe behavior.

[0060] Otherwise, the unsafe behavior identification result is determined as a safe behavior.

[0061] As an optional implementation, triggering an alarm includes: Alert safety supervisors via voice, text message or mobile app.

[0062] Compared with the existing technology, the construction personnel status monitoring and early warning system based on posture tracking has the following significant advantages.

[0063] 1. Strong real-time performance: The high-definition camera of the data acquisition module captures real-time video streams, and uses OpenPose technology to extract the skeleton key points of construction workers in real time. Efficient video processing and posture tracking algorithms enable the system to respond quickly, monitor the movements of construction workers in real time, and promptly identify and alarm unsafe behaviors.

[0064] 2. High precision: The OpenPose technology used in the posture tracking module is based on its high-precision skeleton key point extraction capability, ensuring the accuracy of skeleton key point extraction. Combined with the posture recognition model obtained by the pre-trained machine learning model, the behavior analysis module can accurately identify the actions of construction workers, thereby improving the accuracy of unsafe behavior detection and reducing false positives and false negatives.

[0065] 3. High degree of automation: The system automatically completes the entire process from data collection, skeleton key point extraction to behavior analysis and alarm, reducing dependence on manual intervention and improving monitoring efficiency and response speed. The improvement of automation significantly reduces the errors of manual operation and judgment and improves the reliability of the system.

[0066] 4. Strong scalability: The system adopts a modular design with clear interfaces between modules, which is convenient for function expansion and system upgrade. For example, the monitoring coverage and detection accuracy can be improved by increasing the number of cameras or improving the machine learning model. The existence of the data storage and analysis module enables the system to be continuously optimized and improved based on the accumulated data.

[0067] 5. Improvement of safety management: Using the data storage and analysis module to store and analyze historical data can help discover and improve safety management issues at the construction site and improve the overall safety level. The system can not only monitor and alarm in real time, but also store a large amount of historical data for subsequent analysis and report generation. These data analysis results can help construction managers discover common risk behaviors and inefficient operations, put forward targeted improvement suggestions, and improve the overall safety management level of the construction site.

[0068] In an exemplary embodiment, a construction worker status monitoring and early warning method based on posture tracking is provided, and a construction worker status monitoring and early warning system based on posture tracking is implemented. The construction worker status monitoring and early warning method based on posture tracking includes the following steps.

[0069] Step 101: Capture a live video stream of the construction crew.

[0070] Step 102: Using a computer vision algorithm, extract the skeleton key points of the construction worker from each frame image in the real-time video stream to obtain the coordinates of the skeleton key points of the construction worker.

[0071] Step 103: Input the coordinates of the skeleton key points into the posture recognition model, determine the actions of the construction workers, and use the actions of the construction workers to detect unsafe behaviors and obtain unsafe behavior recognition results; the posture recognition model is obtained by training the machine learning model using a training data set, and the training data set includes the coordinates of multiple skeleton key points and corresponding actions. The unsafe behavior recognition result is an unsafe behavior or a safe behavior.

[0072] Step 104: When the unsafe behavior identification result is an unsafe behavior, an alarm is triggered.

[0073] In an exemplary embodiment, a computer device is provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a construction worker status monitoring and early warning method based on posture tracking.

[0074] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, a construction worker status monitoring and early warning method based on posture tracking is implemented.

[0075] In an exemplary embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements a construction worker status monitoring and early warning method based on posture tracking.

[0076] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 2 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a construction personnel status monitoring and early warning method based on posture tracking is implemented.

[0077] Those skilled in the art will understand that Figure 2The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0078] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0079] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0080] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0081] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the system, method and core ideas of this application; at the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A construction worker status monitoring and early warning system based on posture tracking, characterized in that: The construction personnel status monitoring and early warning system based on posture tracking comprises: a data acquisition module, a posture tracking module, a behavior analysis module and an alarm module connected in sequence; The data acquisition module is used to capture the real-time video stream of the construction personnel and send the real-time video stream to the posture tracking module; The posture tracking module is used to extract the skeleton key points of the construction personnel from each frame image in the real-time video stream by using a computer vision algorithm, obtain the coordinates of the skeleton key points of the construction personnel, and send the coordinates of the skeleton key points to the behavior analysis module; A posture recognition model is deployed in the behavior analysis module, and the behavior analysis module is used to input the coordinates of the skeleton key points into the posture recognition model, determine the actions of the construction personnel, and use the actions of the construction personnel to detect unsafe behaviors, obtain unsafe behavior recognition results, and send the unsafe behavior recognition results to the alarm module; the posture recognition model is obtained by training a machine learning model using a training data set, and the training data set includes the coordinates of multiple skeleton key points and corresponding actions; the unsafe behavior recognition result is an unsafe behavior or a safe behavior; The alarm module is used to trigger an alarm when the unsafe behavior identification result is an unsafe behavior.

2. The construction personnel status monitoring and early warning system based on posture tracking according to claim 1 is characterized in that: The construction personnel status monitoring and early warning system based on posture tracking also includes: a data storage and analysis module; the data acquisition module, the posture tracking module, the behavior analysis module and the alarm module are respectively connected to the data storage and analysis module; The data storage and analysis module is used to store the real-time video stream, the coordinates of the skeleton key points, the unsafe behavior recognition results and the alarm log in the database, and regularly analyze the data stored in the database to generate a safety management report.

3. The construction personnel status monitoring and early warning system based on posture tracking according to claim 1 is characterized in that: The computer vision algorithm is OpenPose technology.

4. The construction personnel status monitoring and early warning system based on posture tracking according to claim 3 is characterized in that: Using OpenPose technology, we can extract the skeleton key points of construction workers from any image and obtain the coordinates of the skeleton key points of construction workers, including: Preprocessing the image to obtain a preprocessed image; the preprocessing includes: adjusting the image size and color channel conversion; The preprocessed image is input into the feature extraction network to obtain multiple feature maps; the number of feature maps is equal to the number of skeleton key points; For each feature map, obtain the heat map corresponding to the key points of the skeleton; Based on the pixel values ​​at all pixels in the heat map of each bone key point, the coordinates of the corresponding bone key point are determined.

5. The construction personnel status monitoring and early warning system based on posture tracking according to claim 1 is characterized in that: The unsafe behavior detection is performed using the actions of the construction workers to obtain the unsafe behavior identification results, including: The actions of the construction workers are compared with the unsafe actions in the predefined unsafe behavior library to obtain the unsafe behavior identification result.

6. The construction personnel status monitoring and early warning system based on posture tracking according to claim 5 is characterized in that: Comparing the actions of the construction workers with the unsafe actions in the predefined unsafe behavior library to obtain the unsafe behavior identification results, including: The similarity between the construction worker's action and the unsafe action is calculated based on the feature vector of the construction worker's action and the feature vector of the unsafe action using the cosine similarity calculation formula; the cosine similarity calculation formula is: ； in, is the similarity between the construction worker's actions and unsafe actions; is the characteristic vector of the construction worker’s action; is the feature vector of the unsafe action; is the dot product; is the magnitude of the vector; If the similarity is greater than a preset similarity threshold, the unsafe behavior recognition result is determined as an unsafe behavior; Otherwise, the unsafe behavior identification result is determined as a safe behavior.

7. The construction personnel status monitoring and early warning system based on posture tracking according to claim 1 is characterized in that: Trigger alarms, including: Alert safety supervisors via voice, text message or mobile app.

8. A construction worker status monitoring and early warning method based on posture tracking, implemented based on the construction worker status monitoring and early warning system based on posture tracking as claimed in any one of claims 1 to 7, characterized in that: The construction personnel status monitoring and early warning method based on posture tracking includes: Capture live video streams of construction crews; Using a computer vision algorithm, extracting skeleton key points of the construction worker from each frame image in the real-time video stream to obtain the coordinates of the skeleton key points of the construction worker; The coordinates of the skeleton key points are input into the posture recognition model to determine the actions of the construction workers, and the actions of the construction workers are used to detect unsafe behaviors to obtain unsafe behavior recognition results; the posture recognition model is obtained by training a machine learning model using a training data set, and the training data set includes the coordinates of multiple skeleton key points and corresponding actions; the unsafe behavior recognition result is an unsafe behavior or a safe behavior; When the unsafe behavior identification result is an unsafe behavior, an alarm is triggered.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the construction personnel status monitoring and early warning method based on posture tracking as described in claim 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the construction personnel status monitoring and early warning method based on posture tracking described in claim 8 is implemented.

Images