Concrete construction process and worker health monitoring system based on computer vision

Abstract

The present invention discloses a computer vision-based concrete construction process and worker health monitoring system, which includes an input module, a computing module, and an output module. The input module collects construction site images in real time and uses embedded devices for real-time processing. The computing module receives data collected by the input module, including the posture of concrete workers, the operating status of equipment, and specific construction task requirements. The system uses improved YOLOv8-D and YOLOv8-P algorithms to accurately detect the posture characteristics of construction equipment and concrete workers. In addition, the system supports SVM to classify and identify construction operation categories, and combines ST-GCN to analyze continuous multi-frame image data to capture the dynamic changes of worker posture and equipment status at the construction site. The functions of the output module include construction process analysis and health monitoring. The system provides feedback on construction status and warning information to the operator in the form of voice broadcast through the human-computer interaction interface, helping them to adjust the operation mode in time to ensure the safety and efficiency of the construction process.

Description

Technical Field

[0001] This invention belongs to the field of building information technology and intelligent construction equipment, specifically relating to a computer vision-based concrete construction process and worker health monitoring system. Background Technology

[0002] With the rapid advancement of infrastructure construction, concrete, as the most commonly used material in building engineering, directly impacts the overall progress and safety of projects through its construction quality and efficiency. However, traditional concrete construction relies on manual operation and monitoring, which is not only inefficient but also prone to quality problems due to errors in pouring and vibration. Furthermore, construction workers maintaining a single posture or repetitive movements for extended periods poses health risks, further affecting construction progress and personnel efficiency. Therefore, automating and intelligentizing the construction process is of great significance for improving construction quality and health monitoring levels.

[0003] In recent years, the rapid development of computer vision and deep learning technologies, such as the application of object detection, pose estimation, and temporal analysis algorithms, has provided new means for intelligent monitoring of construction sites. These technologies make it possible to capture concrete worker movements in real time, monitor equipment operating status, and analyze dynamic construction scenarios. While existing systems can perform basic identification of concrete workers and equipment, they still fall short in terms of real-time process monitoring and accuracy of health monitoring. An automated monitoring system combining YOLOv8, SVM, and ST-GCN algorithms can achieve comprehensive analysis of construction scenarios and dynamic monitoring of personnel health status, contributing to a comprehensive improvement in construction process management and safety assurance. Summary of the Invention

[0004] To address the aforementioned issues, this invention discloses a computer vision-based concrete construction process and worker health monitoring system. This system integrates image processing technology, posture detection, and time-series analysis technology. By analyzing real-time image data and worker posture behavior at the construction site, the system can automatically identify and optimize the concrete construction process, thereby improving construction quality and worker health.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] A computer vision-based concrete construction process and worker health monitoring system includes an input module, a calculation module, and an output module.

[0007] Furthermore, the input module is responsible for receiving and processing data information from the construction site, including personnel posture, equipment operation status and their interactive behavior. Through real-time data collection and transmission, intelligent identification and monitoring of the construction site can be achieved.

[0008] Furthermore, the specific functions of the input module are as follows:

[0009] The input module, as the core of data acquisition, is responsible for the real-time acquisition and transmission of construction site information, providing fundamental data support for subsequent analysis. This module captures construction footage through embedded devices and high-definition cameras, comprehensively recording the status information of concrete workers and equipment, ensuring real-time monitoring of on-site dynamics. Specifically, the input module can capture the physical movements and postures of concrete workers, such as key actions like standing, bending, and raising their hands. Furthermore, the input module can also identify interactions between personnel and equipment, such as the dynamic process of operating a concrete placing boom or mixer, and simultaneously acquire multi-channel data, such as video footage and environmental parameters, ensuring data consistency over time. Through these functions, the input module provides a high-quality data foundation for process monitoring and health management at the construction site.

[0010] The input module supports multiple input methods, can monitor the construction site through image processing technology, and can process the posture and equipment operation data collected on site in combination with artificial intelligence technology.

[0011] Furthermore, the computing module receives data collected by the input module, and classifies and identifies construction operations based on the Support Vector Machine (SVM) classification model and the Spatial-Temporal Graph Convolutional Network (ST-GCN), monitors the construction process, and determines whether an early warning needs to be issued, as detailed below:

[0012] 1. Construction Site Image Processing. The system uses an embedded Jetson Nano camera to capture real-time images of the construction site and employs YOLOv8 image processing technology to detect and locate key targets. This detection process includes three stages: feature extraction, feature fusion, and target prediction.

[0013] (1) In the feature extraction stage, the input image I extracts preliminary features F via the backbone network:

[0014] F = Backbone(I)

[0015] The backbone network extracts spatial and semantic information from images through multi-layer convolutional operations, providing feature support for subsequent detection.

[0016] (2) In the feature fusion stage, the system fuses features at different scales through the feature fusion module to generate a multi-scale feature map F. fusion

[0017] F fusion =Neck(F)

[0018] This step enhances the multi-scale detection capabilities for workers, concrete placing booms, and mixers by combining multi-level information, ensuring detection accuracy in complex construction environments.

[0019] (3) In the target prediction stage, the prediction module outputs the target's category probability P, bounding box parameters B = (x, y, w, h), and confidence score C; the detection result D is represented as:

[0020] D = {(P} i B i C i |i=1,2,...,N}

[0021] Where N is the number of detected targets, P i For the target category, B i The bounding box representing the target contains the center coordinates (x, y) and dimensions (w, h), C i Indicates the confidence score;

[0022] The system further employs a non-maximum suppression algorithm to remove redundant detection boxes, retaining only the target box B with the highest confidence level. max :

[0023]

[0024] Among them, IoU(B i B j ) represents bounding box B i and B j The intersection-union ratio is used to quantify the degree of overlap of the target area. Through the above process, the system achieves accurate detection and positioning of workers, concrete placing booms and mixers, providing real-time monitoring and support for key activities in the construction process, and effectively improving construction efficiency and safety.

[0025] 2. Construction Activity Monitoring. The system uses an SVM classification model to identify the categories of detected actions, such as concrete pouring and vibration operations. The SVM is based on the input posture feature vector X, and uses the hyperplane function f(X) = ω T The X+b method determines the action category and performs dynamic behavior analysis in conjunction with ST-GCN. The system further compares the analysis results with the time requirements of construction procedures; for example, the duration of concrete pouring operations must not exceed 90 minutes, and the time for a single vibration operation should be kept within the range of 5 to 15 seconds. When the system detects that the actual duration of a certain procedure exceeds the specified standard or the operation frequency deviates from the standard value, it will generate an abnormal construction activity warning, reminding construction personnel to adjust operations or optimize the construction rhythm to ensure that the construction task proceeds as planned.

[0026] 3. Health Monitoring. The system calculates the elbow angle and movement frequency of concrete workers in real time based on posture estimation technology. This is achieved by calculating the elbow-shoulder angle θ using the formula... Determine if the movement exceeds a health threshold. Points A, B, and C represent the positions of the shoulder, elbow, and wrist, respectively, and the vector... and These represent the directions from shoulder to elbow and from shoulder to wrist, respectively. ST-GCN captures the interaction between concrete workers and equipment through graph convolution operations, mapping multi-frame data into temporal features to further assess whether the duration and frequency of operations comply with specifications. When the elbow angle θ > 150° or the hand-raising frequency f > 20 times / minute, the system generates a health warning signal, reminding the concrete worker to adjust their operation through voice broadcast and interface prompts to avoid health problems caused by repetitive movements.

[0027] 4. Construction Activity Monitoring and Feedback. The system dynamically tracks the actual progress of construction tasks. When the operation time exceeds a preset threshold or unhealthy operation is detected, the system issues an alert and displays specific abnormal information through the human-computer interaction interface, ensuring the improvement of construction quality and efficiency.

[0028] Furthermore, the output module provides real-time feedback to construction operators through a human-computer interaction interface and voice broadcast system, ensuring that they can adjust their operations in a timely manner, guaranteeing the safety and efficiency of construction, and reducing health risks caused by improper operation.

[0029] The method of using the computer vision-based concrete construction process and worker health monitoring system of the present invention includes the following steps:

[0030] 1. On-site scene analysis: The input module uses computer vision technology to process the concrete construction site video acquired by the embedded device, including posture recognition and image analysis, to extract core information about the concrete construction, including but not limited to personnel location, equipment status, and key indicators of construction activities.

[0031] 2. Construction Status Identification and Analysis: The system utilizes YOLOv8 and SVM models, combined with ST-GCN, to perform time-series analysis, monitor the actions of personnel and the status of equipment at the construction site in real time, generate detailed construction status information, and detect abnormal behavior to ensure that every aspect of the construction process is accurately monitored.

[0032] 3. Early warning feedback: The system determines whether an early warning needs to be issued based on the calculation results of the construction process or the health status of the workers. The early warning information is provided to the operator through intuitive voice broadcast and interface display to ensure the smooth progress of the construction process and protect the health of the workers.

[0033] The beneficial effects of this invention are:

[0034] 1. Immediacy and convenience. Through embedded devices, managers can obtain concrete construction status anytime, anywhere via this system, improving management flexibility and response speed.

[0035] 2. Accuracy. By combining the SVM model, YOLOv8, and ST-GCN, the system can monitor the construction site in real time, ensuring the automation and intelligence of the construction process and health monitoring. At the same time, the system's high-precision detection capability can accurately capture dynamic changes in human-computer interaction, maintaining high efficiency and reliability even in complex and ever-changing construction environments, thereby significantly improving the accuracy of abnormal posture recognition and effectively avoiding misunderstandings or information transmission distortions caused by human factors.

[0036] 3. Health Protection. The system can monitor workers' health status in real time, especially detecting repetitive movements such as frequent arm raising, which helps prevent occupational diseases and protect workers' health.

[0037] 4. Improved Efficiency. The system provides real-time feedback and guidance, helping operators adjust their operations promptly, optimize construction processes, and improve work efficiency. This increased efficiency not only accelerates construction progress but also reduces costs caused by delays and rework, saving time and resources for the project.

[0038] 5. 24 / 7 Service. This invention provides 24 / 7 service, ensuring the continuity and timeliness of construction management. Even at night or in inclement weather conditions, the system functions normally, providing managers with the necessary information. This round-the-clock monitoring and feedback mechanism provides additional security for construction projects. Attached Figure Description

[0039] Figure 1 This is a framework diagram for implementing the present invention.

[0040] Figure 2 This is a diagram of a worker's posture. Detailed Implementation

[0041] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0042] As shown in the figure, the computer vision-based concrete construction progress and worker health monitoring system of the present invention includes:

[0043] 1. Input module

[0044] The input module is designed to adapt to the complex environment of construction sites, integrating a camera and embedded devices such as a Jetson Nano. The system captures real-time images of the construction site through a high-definition camera and uses the YOLOv8 algorithm for worker posture recognition and equipment detection. The main function of this module is to extract key construction information from the images, such as workers, equipment, and their spatial coordinates.

[0045] 2. Calculation Module

[0046] The computation module utilizes SVM combined with YOLOv8-P and YOLOv8-D algorithms to perform real-time classification and identification of construction tasks, monitor the duration of concrete pouring and vibration operations, and perform time-series data analysis using ST-GCN. Specific functions are as follows:

[0047] a. Equipment Detection. The calculation module uses the YOLOv8 model to detect equipment and identify construction scenes in images of the construction site, providing crucial data support for subsequent construction activity analysis and operational plan development. The system uses YOLOv8 to detect targets such as concrete workers, concrete placing booms, and vibratory compactors, extracting their category labels and location coordinates, including center point coordinates (x, y) and dimensional parameters (w, h).

[0048] First, during the model training phase, the system optimizes the YOLOv8 model using a binary cross-entropy loss function and dynamically adjusts the learning rate and momentum parameters using the Adam optimizer to ensure efficient model convergence and improve detection accuracy. Then, during the inference phase, the input construction site image extracts feature maps through the YOLOv8 backbone network, which are then integrated at multiple scales via a feature fusion module. Finally, the prediction module generates the target's category and location information.

[0049] Next, in the post-processing stage, the system uses non-maximum suppression to remove redundant detection boxes, retaining only the boxes with the highest confidence, thereby improving the reliability and accuracy of the detection results. Finally, the system outputs the category labels of workers and equipment and their corresponding spatial location information as structured data to support subsequent construction task analysis and scheme optimization.

[0050] b. Construction Activity Monitoring. Through multi-level analysis of dynamic data from the construction site, the system achieves process identification and status monitoring. First, the system uses a YOLOv8 model to detect key targets in the construction task, including the status and location of concrete workers and equipment such as concrete placing booms and vibrators, extracting their category labels and location information. Then, the system combines SVM to classify the detected process categories, clarifying different operation types such as concrete pouring and vibration. SVM makes judgments based on posture feature vectors, using the decision function f(X) = w T X+b determines the operation category.

[0051] To further capture the dynamic characteristics of the work processes, the system introduces ST-GCN for time-series analysis of continuous frame data. ST-GCN models the interaction behavior of concrete workers and equipment, providing in-depth analysis of key indicators such as the duration and frequency of work processes. For example, the system can monitor whether the duration of a single concrete pouring operation exceeds 90 minutes, and whether the duration of a single vibration operation remains within the specified range of 5 to 15 seconds. When the continuous duration of equipment use exceeds safety limits or non-compliant worker behavior is detected, the system generates an anomaly warning. Simultaneously, the structured work process information generated by the system not only assesses the status of construction tasks but also provides a basis for optimizing construction plans, further improving construction efficiency.

[0052] c. Health Early Warning. By analyzing the posture information and movement characteristics of concrete workers in real time, the system achieves dynamic monitoring and early warning of their health status. The system first calculates key indicators based on the concrete workers' posture data, such as elbow angle and movement frequency. The formula for calculating the elbow angle θ is as follows: (Formula omitted for brevity) in, and These represent the vectors from the shoulder to the elbow and from the shoulder to the wrist, respectively. When the elbow angle θ > 150° or the arm raising frequency f > 20 times / minute, the system determines it as a high-risk movement and generates a health warning signal.

[0053] The system combines ST-GCN to perform time-series analysis of concrete workers' dynamic behavior, assessing the duration of actions and the frequency of repetitive operations. When a posture or frequency exceeds a set threshold, such as maintaining a single action for a long time or raising the hand too frequently, the system reminds the concrete worker to adjust their operation in a timely manner through voice broadcasts and interface prompts, avoiding muscle strain or fatigue caused by overwork.

[0054] 3. Output module

[0055] The output module is responsible for providing the administrator with the construction activity inspection results and worker health monitoring results generated by the calculation module, via text and voice. Specific functions are as follows:

[0056] a. Text Display. The system will display construction procedures and worker health check results in text format on the user interface. Administrators can directly view abnormal construction activities and warning messages.

[0057] b. Voice broadcast: To facilitate information access for concrete construction workers when they cannot view the screen during operation, the system supports voice broadcast via speaker, providing real-time feedback on the construction process and health warning prompts.

[0058] System physical architecture:

[0059] 1. User Interface Layer

[0060] User interface: Provides an intuitive and easy-to-use interface for managers to view the construction process and warning messages.

[0061] Input devices include embedded devices such as Jetson Nano cameras, used to receive images from the construction site.

[0062] Output devices, such as displays and speakers, are used to show process progress or provide voice broadcasts of health reminders.

[0063] 2. Infrastructure Layer

[0064] Embedded device Jetson Nano camera: used to acquire images of the construction site and perform preliminary processing.

[0065] Servers and storage devices: provide computing resources and data storage space to ensure stable system operation.

[0066] Network equipment: Ensures smooth data communication between various components and supports real-time data transmission and feedback.

[0067] It should be noted that the above content merely illustrates the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, various improvements and modifications can be made without departing from the principle of the present invention, and all such improvements and modifications fall within the scope of protection of the claims of the present invention.

Claims

1. A computer vision-based concrete construction process and worker health monitoring system, characterized in that: It includes an input module, a calculation module, and an output module. The computing module receives data collected by the input module, and classifies and identifies construction operations based on the SVM classification model and ST-GCN, monitors the construction process, and determines whether an early warning needs to be issued, as follows: a. Understanding construction scenes This system is based on the embedded device Jetson Nano. It captures real-time images of the construction site through a camera and uses YOLOv8 image processing technology to detect and locate key targets. The entire detection process is divided into three stages: feature extraction, feature fusion, and target prediction. In the feature extraction stage, the backbone network extracts spatial and semantic information of the image through convolutional kernel combination, activation function and pooling layer, providing basic support for detection; In the feature fusion stage, the feature pyramid network enhances the multi-scale detection capability of workers, concrete placing machines, and mixers, ensuring detection accuracy in complex environments. In the target prediction stage, the prediction module outputs the target category, bounding box, and confidence score, and removes redundant boxes using a non-maximum suppression algorithm, retaining only the result with the highest confidence score. b. Construction process monitoring The system uses an SVM classification model to identify concrete pouring and vibration operations based on detected action categories. SVM determines the action category based on the input posture feature vector X and combines it with ST-GCN to perform dynamic temporal analysis of construction behavior. The system extracts the duration and frequency of actions and compares the analysis results with the time requirements of the construction process. Specifically, the duration T of concrete pouring operations should satisfy T1≤90min, and the duration T2 of a single vibration operation should satisfy 5s≤T2≤15s. Otherwise, the system will trigger an abnormal construction activity warning, reminding construction personnel to adjust their operating strategies. Warning information is displayed through voice prompts and a visual interface, helping construction personnel optimize the construction rhythm and ensure tasks are completed as planned, while simultaneously improving construction safety and quality management. c. Health monitoring The system can monitor concrete workers' posture information in real time, including elbow extension and frequent hand raising behaviors. Based on the YOLOv8-Pose algorithm, the system first extracts the key point coordinates of the concrete worker, including the positions of the shoulder, elbow, and wrist; by calculating the elbow angle θ, the system determines whether the elbow is in an extended or raised state; specifically, the formula for calculating the angle θ is... Points A, B, and C represent the positions of the shoulder, elbow, and wrist, respectively, and the vector... and These represent the directions from shoulder to elbow and from shoulder to wrist, respectively; when the elbow angle θ > 150°, the system determines that the elbow is in an overstretched state. In addition, the YOLOv8-Pose algorithm records every time a worker raises or lowers their hand, thereby counting the number of times a concrete worker raises their hand f per minute. Through the ST-GCN algorithm, if the frequency of the movement f exceeds the preset health threshold of 20 times / minute, the system will trigger a health warning to remind the concrete worker to take a break and prevent fatigue or muscle strain caused by overwork.

2. The computer vision-based concrete construction process and worker health monitoring system according to claim 1, characterized in that: The input module is responsible for receiving data from the construction site, including image data collected by the camera and relevant environmental parameter data, and transmitting this data to the computing module for further processing and analysis.

3. The computer vision-based concrete construction process and worker health monitoring system according to claim 2, characterized in that: The specific function of the input module is that the system uses the YOLOv8 algorithm to identify key entities and location information in the construction task, and converts this information into structured query data for further analysis by the calculation module.

4. The computer vision-based concrete construction process and worker health monitoring system according to claim 2, characterized in that: The input module collects images and videos of the construction site through a camera and processes the posture and equipment operation data collected on site using artificial intelligence technology.

5. The computer vision-based concrete construction process and worker health monitoring system according to claim 1, characterized in that: The output module implements multi-level construction status feedback functions through a human-computer interaction interface, including three main aspects: health warning, operation guidance, and risk alerts. First, for the health risk monitoring of concrete workers, the system dynamically assesses the operation status by detecting elbow angle and movement frequency. When the elbow angle exceeds the set range or the movement frequency exceeds the health threshold, the system immediately generates a warning and transmits the warning information to the operator through voice broadcast and visual interface, prompting the operator to rest or adjust the operation mode to avoid occupational health problems caused by long-term repetitive operation. The system's health monitoring and feedback mechanism is data-driven. By accurately monitoring elbow extension and frequent hand raising behaviors, combined with real-time voice and visual feedback, it effectively improves the level of intelligence in health management at construction sites. At the same time, the mechanism uses a dual early warning system of voice broadcasts and interface prompts to ensure accurate information transmission while ensuring that operators can respond quickly and adjust non-standard postures or movements. Ultimately, the system not only achieves health risk prevention and control during construction but also provides reliable support for improving the overall operational safety of concrete workers.

6. A method for monitoring concrete construction process and worker health based on computer vision, characterized in that: Includes the following steps: a. On-site scene analysis: Video data from the construction site is collected through embedded devices. The input module uses computer vision technology to preprocess the images, including target detection and key point extraction. The system uses the YOLOv8 algorithm to identify concrete workers, concrete placing booms, and mixers at the construction site and records their position coordinates and status. At the same time, it combines pose estimation technology to capture the limb movement information of concrete workers, providing multi-dimensional data support for subsequent analysis. b. Construction Status Identification and Analysis: The system comprehensively utilizes YOLOv8 target detection, SVM classification model, and ST-GCN temporal analysis model to identify and analyze the dynamic behavior of the construction site. First, the SVM model is used to classify the detected concrete pouring and vibration actions. Then, ST-GCN monitors the interaction behavior of personnel and equipment and the changing trends of continuous actions based on the temporal features of multi-frame images. The system transforms the analysis results into construction status information, including construction procedures, personnel operation status, and equipment operating status, providing support for real-time monitoring. c. Health monitoring The system monitors concrete workers' posture information in real time, including elbow extension and frequent hand raising behaviors. Based on the YOLOv8-Pose algorithm, the system first extracts the key point coordinates of the concrete workers, including the positions of the shoulder, elbow, and wrist. By calculating the elbow angle θ, the system determines whether the elbow is in an extended or raised state. Specifically, the formula for calculating the angle θ is... Points A, B, and C represent the positions of the shoulder, elbow, and wrist, respectively, and the vector... and These represent the directions from shoulder to elbow and from shoulder to wrist, respectively; when the elbow angle θ > 150°, the system determines that the elbow is in an overstretched state. In addition, the YOLOv8-Pose algorithm records every time a worker raises or lowers their hand, thereby counting the number of times a concrete worker raises their hand f per minute. Through the ST-GCN algorithm, if the frequency of the movement f exceeds the preset health threshold of 20 times / minute, the system will trigger a health warning to remind the concrete worker to take a break and prevent fatigue or muscle strain caused by overwork. d. Early warning feedback: Based on the analysis results, the system assesses the health status and operational compliance of concrete workers; When abnormal postures are detected, such as an elbow angle exceeding 150°, a hand-raising frequency exceeding 20 times per minute, or a single concrete pouring operation lasting more than 90 minutes, or a single vibration operation time not being kept within the specified range of 5 to 15 seconds, the system will generate an early warning signal. The early warning information will be delivered to the concrete workers in real time via voice broadcast, and the specific problems will be displayed visually on the interface, including the type, frequency, and duration of the abnormal movement. Operators can adjust their operating behavior, optimize their operating rhythm, or take appropriate breaks based on the feedback information to ensure that the construction task is carried out as required, while reducing occupational health risks.

Images