Engineering safety quality supervision patrol data processing method and system

Abstract

The invention provides an engineering safety quality supervision patrol data processing method and system, and relates to the field of data processing. In the method, a construction plan corresponding to a target time period is obtained, and the construction plan comprises first constructor information and preset construction step information; in the target time period, a first image sent by AR glasses is obtained, and the user wears the AR glasses; analyzing the first image to obtain second constructor information; if it is determined that the first constructor information is consistent with the second constructor information, obtaining a second image sent by the AR glasses; identifying the second image to obtain a real-time construction step; and generating real-time construction prompt information corresponding to the real-time construction steps according to the preset construction step information. By implementing the technical scheme provided by the invention, the supervision efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a method and system for processing engineering safety and quality supervision inspection data. Background Art

[0002] In today's vast field of construction engineering, supervision is an indispensable part of ensuring project quality, and its importance is becoming increasingly prominent. By using advanced data monitoring and analysis technology, supervision personnel implement in-depth and meticulous supervision of the entire process of engineering construction, aiming to pre-identify and effectively solve potential problems in the construction process, ensure that the project strictly follows the established quality standards, and ultimately achieve high-quality delivery.

[0003] However, in the face of the continuous expansion of the scale of construction projects and their increasing complexity, the traditional supervision model that relies on manual visual inspection and experience judgment is facing severe challenges. Under this model, the workload of supervisors who rely on traditional drawings and verbal guidance is becoming increasingly heavy, and due to differences in individual experience levels, the supervision efficiency is relatively low.

[0004] Therefore, there is an urgent need for a method and system for processing engineering safety and quality supervision inspection data. Summary of the invention

[0005] The present application provides a method and system for processing engineering safety and quality supervision inspection data, which is convenient for improving the supervision efficiency of construction supervision.

[0006] In a first aspect of the present application, a method for processing engineering safety and quality supervision inspection data is provided, the method comprising: obtaining a construction plan corresponding to a target time period, the construction plan comprising first construction personnel information and preset construction step information; in the target time period, obtaining a first image sent by an AR glasses, the user wearing the AR glasses; analyzing the first image to obtain second construction personnel information; if it is determined that the first construction personnel information and the second construction personnel information are consistent, obtaining a second image sent by the AR glasses; identifying the second image to obtain real-time construction steps; and generating real-time construction prompt information corresponding to the real-time construction steps according to the preset construction step information.

[0007] By adopting the above technical solution, by automatically acquiring and analyzing the image of the construction worker wearing AR glasses, the system can instantly confirm the identity of the construction worker (i.e., the consistency of the first construction worker information and the second construction worker information), so as to quickly enter the next step of operation. This reduces the time for manual verification and confirmation and improves the efficiency of the overall construction process. Confirming the identity of the construction worker is an important part of construction safety management. This method uses technical means to ensure that only authorized and confirmed construction workers can perform specific steps of construction, effectively preventing safety hazards caused by misoperation by non-professionals. The system can automatically generate corresponding real-time construction prompt information based on the preset construction step information and the real-time identified construction steps (the second image sent by AR glasses). This instant and accurate feedback mechanism helps construction workers to follow the correct steps for construction, reduce erroneous operations, and improve construction quality. The introduction of AR glasses provides construction workers with a new way of working. Through intuitive visual prompts and real-time feedback, construction workers can complete construction tasks more easily and accurately, while reducing misunderstandings and communication barriers that may be caused by traditional paper drawings or verbal instructions, and improving supervision efficiency.

[0008] Optionally, obtaining the construction plan corresponding to the target time period specifically includes: receiving plan data uploaded by the user; preprocessing the plan data to obtain the construction plan corresponding to the target time period, the preprocessing including denoising, filtering, correspondence extraction and normalization.

[0009] By adopting the above technical solutions, the noise or erroneous information that may exist in the planning data can be removed to ensure that the basic data for subsequent analysis is accurate and reliable. This is crucial to ensure the accurate execution of the construction plan. Through filtering, fluctuations or outliers in the data can be smoothed, further improving the accuracy and consistency of the data. Construction plan information directly related to the target time period, such as construction personnel arrangements, construction steps, required materials, etc., can be extracted from complex planning data. This step makes subsequent analysis and decision-making more focused and efficient. Converting data to the same scale or standard facilitates unified processing and analysis of data from different sources or formats. This helps to enhance the compatibility and comparability of data and facilitates cross-project or cross-departmental collaboration. The preprocessing step can enhance the system's tolerance for bad data and reduce system errors or failures caused by data problems. This helps to improve the stability and reliability of the entire construction management system. Through preprocessing, redundant data can be removed and the amount of data to be processed later can be reduced, thereby improving the speed and efficiency of data processing. This is particularly important for real-time acquisition and analysis of construction plan information.

[0010] Optionally, obtaining the first image sent by the AR glasses in the target time period, and the user wearing the AR glasses, specifically includes: determining whether the target time period includes the current moment; if it is determined that the target time period includes the current moment, controlling the AR glasses to capture images in real time to obtain the first image; if it is determined that the target time period does not include the current moment, searching from multiple images captured by the AR glasses and obtaining the image duration according to the target time period; determining the first image based on the image duration and in combination with a preset corresponding relationship, wherein the preset corresponding relationship is a corresponding relationship between image duration and image.

[0011] By adopting the above technical solution, when the target time period includes the current moment, the system can control the AR glasses to collect images in real time to ensure that the latest and real-time construction scene information is obtained. This real-time performance is essential for timely monitoring of construction progress and discovering potential problems. At the same time, the system is also flexible and can handle situations where the target time period does not include the current moment, and meet the needs by searching historical image data. In the case where the target time period does not include the current moment, the system does not simply select the nearest or farthest image, but accurately searches for images that match the target time period based on the image duration and the preset correspondence. This helps to ensure that the acquired images are highly relevant to specific stages or time points in the construction plan, and improves the accuracy of subsequent analysis and decision-making. Through the preset correspondence, the system can quickly locate images that match the target time period, avoiding the tedious process of searching one by one in a large number of historical images, and improving the efficiency of data retrieval.

[0012] Optionally, the analyzing the first image to obtain the second construction worker information specifically includes: using a preset analysis model to extract facial features and identity features from the first image; inputting the facial features and the identity features into the preset analysis model respectively to obtain a first recognition result and a second recognition result; obtaining the second construction worker information based on the first recognition result and the second recognition result, the second construction worker information including the construction worker's identity information, position information and authorization information related to the construction task.

[0013] By adopting the above technical solution, by extracting facial features and identity features (such as work permits, logos on helmets, etc.) from the first image and inputting them into the preset analysis model respectively, the system can comprehensively consider multiple factors for identification, thereby improving the accuracy of the identification results. This multiple verification mechanism reduces the possibility of misidentification or missed identification. The second construction personnel information obtained includes not only the identity information of the construction personnel (such as name, work number, etc.), but also job information and authorization information related to the construction task. This information provides a comprehensive perspective for construction management, helps to ensure that construction personnel have the corresponding qualifications and permissions, thereby improving the safety and compliance of construction. The entire process is highly automated and intelligent, reducing the need for manual intervention. The preset analysis model can process image data quickly and accurately, and output the required information, improving work efficiency and accuracy. Since the image is collected and analyzed in real time, the system can immediately obtain the latest information of the construction personnel, providing real-time feedback for construction management. This helps to promptly discover and deal with potential problems, such as unauthorized personnel entering the construction area. By verifying the identity and permissions of the construction personnel, the system can ensure that only authorized personnel can perform specific construction tasks. This helps to prevent unauthorized operations and potential safety risks.

[0014] Optionally, the identifying the second image to obtain the real-time construction steps specifically includes: extracting image frames from the second image to obtain multiple image frames; arranging and combining the multiple image frames according to a preset time sequence to obtain a set of images to be identified; using a preset analysis model to extract behavioral features from the multiple image frames in the set of images to be identified to obtain a behavior feature set; performing feature comparison between the behavior feature set and a preset behavior feature set contained in a preset construction step corresponding to a preset construction task to obtain the real-time construction steps.

[0015] By adopting the above technical solution, through image frame extraction, the system can capture every subtle change in the second image and form multiple image frames. This meticulous capture capability helps to more accurately analyze the behavioral characteristics of construction workers. Arrange and combine multiple image frames according to a preset time sequence to obtain a set of images to be identified. This process takes into account the temporal sequence of construction actions and helps the system better understand the coherence and sequence of construction steps. The preset analysis model can automatically extract the behavioral characteristics of construction workers from the set of images to be identified to form a set of behavioral characteristics. This process reduces the need for manual intervention and improves the efficiency and accuracy of data processing. By comparing the extracted behavioral feature set with the preset behavioral feature set in the preset construction steps corresponding to the preset construction task, the real-time construction steps currently being performed by the construction workers can be accurately identified. This precise comparison mechanism helps to reduce misjudgments and missed judgments.

[0016] Optionally, generating real-time construction prompt information corresponding to the real-time construction step based on the preset construction step information specifically includes: extracting the preset construction prompt information corresponding to the preset construction step information from a preset database, wherein the preset database pre-stores the correspondence between the preset construction step information and the preset construction prompt information; marking the preset construction prompt information to the real-time construction step to obtain a screen to be displayed, and the screen to be displayed includes the real-time construction prompt information.

[0017] By adopting the above technical solution, by extracting the preset construction prompt information corresponding to the real-time construction steps from the preset database, the system can ensure that the guidance information provided is accurate and targeted. This accuracy helps construction personnel better understand the construction requirements and reduce operational errors and misunderstandings. The instant display of real-time construction prompt information enables construction personnel to immediately obtain detailed guidance on the current construction steps. This helps them to perform construction tasks quickly and accurately, reduce delays and duplication of work caused by waiting for guidance or misunderstandings, and thus improve construction efficiency. Traditional construction guidance often requires frequent communication between on-site managers and construction personnel. By automatically generating and displaying real-time construction prompt information, this communication demand can be reduced, communication costs can be reduced, and misunderstandings and errors caused by poor communication can be reduced. Real-time construction prompt information often contains safety-related precautions and warning information. By displaying this information to construction personnel in a timely manner, they can be reminded to pay attention to potential safety hazards and take necessary protective measures, thereby improving safety during the construction process. The correspondence between the preset construction step information and the preset construction prompt information stored in the preset database provides a standardized framework for construction management. This standardization helps to ensure the standardization and consistency of the construction process and improve the overall construction management level.

[0018] Optionally, before adopting the preset analysis model to extract facial features and identity features from the first image, the preset analysis model is trained; the training of the preset analysis model specifically includes: obtaining training information; inputting the training information into an adaptive feature fusion network for training to obtain a first training result; superimposing and standardizing the first training result and the training information to obtain a second training result; inputting the second training result into the adaptive feature fusion network for processing to obtain a third training result; superimposing and standardizing the third training result and the second training result until the training information similarity matrix is ​​output, and the training information similarity matrix satisfies a preset logistic regression condition.

[0019] By adopting the above technical solution, the model can gradually learn more complex and sophisticated feature representations by inputting the training information into the adaptive feature fusion network for iterative training multiple times, and performing superposition and standardization after each iteration. This gradually refined training method helps to improve the recognition accuracy of the model for facial features and identity features. The design of the adaptive feature fusion network enables the model to automatically adjust the weights between different features, so as to better adapt to the facial and identity features under different scenes and lighting conditions. This flexibility enables the model to have stronger generalization ability in practical applications and can handle more diverse input data. Through iterative training, the model can gradually approach the optimal solution. Each iteration is adjusted based on the results of the previous one, thus avoiding repeated calculations from the beginning. This training method not only reduces the waste of computing resources, but also significantly improves the training efficiency. During the training process, by introducing the similarity matrix and the preset logistic regression condition as the criteria for stopping training, it can ensure that the model stops training after reaching a certain accuracy, avoiding the problems of overfitting and training instability. This mechanism helps to maintain the stability and reliability of the model.

[0020] In a second aspect of the present application, a construction safety and quality supervision inspection data processing system is provided, the data processing system comprising an acquisition module and a processing module, wherein the acquisition module is used to acquire a construction plan corresponding to a target time period, the construction plan comprising first construction personnel information and preset construction step information; the acquisition module is also used to acquire a first image sent by an AR glasses, wherein the user is wearing the AR glasses, during the target time period; the processing module is used to analyze the first image to obtain second construction personnel information; the processing module is also used to acquire a second image sent by the AR glasses if it is determined that the first construction personnel information and the second construction personnel information are consistent; the processing module is also used to identify the second image to obtain real-time construction steps; the processing module is also used to generate real-time construction prompt information corresponding to the real-time construction steps according to the preset construction step information.

[0021] In the third aspect of the present application, an electronic device is provided, which includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device performs the method described above.

[0022] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, the method described above is executed.

[0023] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: By automatically acquiring and analyzing the images of construction workers wearing AR glasses, the system can instantly confirm the identity of the construction workers (i.e., the consistency between the first construction worker information and the second construction worker information), and quickly proceed to the next step. This reduces the time for manual verification and confirmation, and improves the efficiency of the overall construction process. Confirming the identity of construction workers is an important part of construction safety management. This method uses technical means to ensure that only authorized and confirmed construction workers can perform specific steps of construction, effectively preventing safety hazards caused by misoperation by non-professionals. The system can automatically generate corresponding real-time construction prompt information based on the preset construction step information and the real-time identified construction steps (the second image sent by AR glasses). This instant and accurate feedback mechanism helps construction workers to follow the correct steps for construction, reduce erroneous operations, and improve construction quality. The introduction of AR glasses provides construction workers with a new way of working. Through intuitive visual prompts and real-time feedback, construction workers can complete construction tasks more easily and accurately, while reducing misunderstandings and communication barriers that may be caused by traditional paper drawings or verbal instructions, and improving supervision efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of a method for processing engineering safety and quality supervision inspection data provided in an embodiment of the present application.

[0025] Figure 2 Another flowchart of a method for processing engineering safety and quality supervision inspection data provided in an embodiment of the present application.

[0026] Figure 3 A module diagram of a project safety and quality supervision inspection data processing system provided in an embodiment of the present application.

[0027] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0028] Explanation of the reference numerals: 31, acquisition module; 32, processing module; 41, processor; 42, communication bus; 43, user interface; 44, network interface; 45, memory. DETAILED DESCRIPTION

[0029] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0030] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "for example" or "for example" is intended to present related concepts in a specific way.

[0031] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0032] In today's grand stage of construction engineering, the role of supervision as the mainstay of safeguarding project quality has become increasingly strategically valuable. The supervision team is actively embracing the power of technology, using cutting-edge data monitoring and in-depth analysis technology to conduct all-round, high-precision dynamic supervision of engineering projects. This transformation aims to proactively discover and efficiently resolve various risks and problems hidden in the construction process, ensuring that each project can accurately meet and exceed the established quality standards, thereby laying a solid foundation for high-quality delivery.

[0033] However, with the continuous expansion of the scale of construction projects and the sharp increase in their inherent complexity, the traditional supervision model, which is highly dependent on manual visual inspection and experience accumulation, is facing unprecedented challenges. Under this model, supervisors not only need to deal with a large amount of traditional drawing information, but also need to provide frequent verbal guidance, and the workload has risen sharply. More importantly, since the experience level of each supervisor is different, this difference has limited the overall efficiency and consistency of the supervision work to a certain extent, making the traditional method incapable of dealing with large and complex projects.

[0034] In order to solve the above technical problems, this application provides a method for processing engineering safety quality supervision inspection data, referring to Figure 1 , Figure 1 A flowchart of a method for processing engineering safety quality supervision inspection data provided in an embodiment of the present application. The data processing method is applied to an AR glasses processor and includes steps S110 to S160, which are as follows: S110: Obtain a construction plan corresponding to the target time period, where the construction plan includes first construction personnel information and preset construction step information.

[0035] Specifically, the AR glasses processor is a high-performance computer used to process network requests, store and manage data, etc. In this scenario, it is responsible for providing construction plan information upon request. The target time period refers to a time range specified by the user or system for filtering or locating a specific construction plan. This time period can be a day, a week, a month, or even a more specific start and end time point. The construction plan is a collection of files or data for planning, arranging and deploying upcoming construction activities. It details key information such as the goals, steps, required resources, and time schedule of the construction. The first construction personnel information refers to the construction personnel information involved in the construction plan, including but not limited to the name, work number, position, responsibilities, contact information, etc. of the construction personnel. This information is crucial for coordinating construction activities and ensuring construction quality and safety. The preset construction step information refers to the construction steps and processes listed in detail in the construction plan, including the specific content, execution order, required time, precautions, etc. of each step. This information is an important basis for guiding the orderly progress of construction activities.

[0036] In a possible implementation, obtaining a construction plan corresponding to a target time period specifically includes: receiving plan data uploaded by a user; preprocessing the plan data to obtain a construction plan corresponding to the target time period, the preprocessing including denoising, filtering, corresponding relationship extraction, and normalization.

[0037] Specifically, the user uploads data containing construction plan information to the AR glasses processor in some way (such as a web form, file upload interface, etc.). This data may be stored in text, Excel spreadsheets, JSON or XML formats, and contains construction plan information for multiple time periods. Preprocessing is an important stage of data cleaning and preparation, which aims to improve data quality and make it more suitable for subsequent analysis and processing. For construction plan data, preprocessing may include multiple steps, such as denoising, filtering, correspondence extraction, and normalization. Denoising is the removal of noise or irrelevant information in the data, such as invalid characters, duplicate entries, or incorrect dates. This helps ensure the accuracy and consistency of the data. In some cases, it may be necessary to smooth the data or remove high-frequency noise through filtering techniques, but this may not be necessary in the preprocessing of construction plan data unless the data contains some time series information that needs to be smoothed. Extract the construction plan information corresponding to the target time period from the raw data. This may involve parsing the data structure (such as rows and columns of Excel spreadsheets), identifying and extracting the construction steps, personnel information, etc. that match the target time period. Convert data of different dimensions or ranges to the same dimension or range for subsequent analysis and processing. In the construction plan data, normalization may not be necessary, but if the data contains some values ​​that need to be compared (such as time, cost, etc.), normalization may be required to ensure the fairness of the comparison. After preprocessing, the system will generate one or more construction plans corresponding to the target time period. These plans contain the construction steps, personnel information, etc. that have been cleaned and organized, and can be directly used for the guidance and management of construction activities.

[0038] For example, suppose a user uploads an Excel file that contains a monthly construction plan from January to December 2023. There are multiple worksheets in the file, each corresponding to a month, and each row in the worksheet represents a construction step, including information such as the step name, execution date, and required personnel. The user selects and uploads this Excel file through the web interface of the project management AR glasses processor. The AR glasses processor first reads the Excel file and identifies the worksheet structure. For each worksheet (i.e., each month), the AR glasses processor performs denoising operations to remove invalid characters, blank lines, etc. Then, the AR glasses processor extracts the construction step information corresponding to the target time period (such as April 2023) based on the date column in the worksheet. If the data contains some values ​​that need to be compared (which may not be common in this example), the AR glasses processor may also perform normalization processing. But in this scenario, the main task is to extract corresponding relationships. After preprocessing, the AR glasses processor generates a construction plan containing all construction step information for April 2023. This plan may be displayed in a tabular form on a web page, or returned to the caller in a format such as JSON, XML, etc. for further use or display.

[0039] S120. Acquire a first image sent by the AR glasses in a target time period, where the user is wearing the AR glasses.

[0040] Specifically, AR glasses are a wearable device that can superimpose digital information into the user's field of view, thereby providing an augmented reality experience. They are equipped with cameras, sensors, and displays that can capture images of the real world and combine them with virtual information. In this scenario, the user is wearing AR glasses and using them to observe the surrounding environment for engineering supervision. AR glasses not only provide users with an enhanced visual experience, but may also transmit the images seen by the user to the AR glasses processor in real time. During the target time period, the AR glasses processor receives the first image from the AR glasses. This image may be a real-time capture of the user's current field of view, or it may be a static image captured under specific conditions (such as when the user triggers the photo function). Among them, the user is an engineering supervisor.

[0041] In a possible implementation, in a target time period, a first image sent by the AR glasses is obtained, and the user wears the AR glasses, specifically including: determining whether the target time period includes the current moment; if it is determined that the target time period includes the current moment, controlling the AR glasses to capture images in real time to obtain the first image; if it is determined that the target time period does not include the current moment, searching from multiple images captured by the AR glasses and obtaining the image duration according to the target time period; determining the first image based on the image duration and a preset corresponding relationship, where the preset corresponding relationship is a corresponding relationship between image duration and images.

[0042] Specifically, the AR glasses processor needs to determine whether the current moment is within the target time period specified by the user or the system. This is to ensure that the image acquisition or retrieval process is started only when necessary. If the current moment is indeed within the target time period, the AR glasses processor will control the AR glasses to perform real-time image acquisition. This means that the camera of the AR glasses will be activated to capture the image in the user's current field of view and send this image as the first image to the AR glasses processor. If the current moment is not within the target time period, but the AR glasses processor still needs to obtain images within the time period, it will take another approach. The AR glasses processor will search for images that match the target time period based on the timestamp from multiple images that the AR glasses have previously captured and stored. In this process, the AR glasses processor calculates the "image duration" of each image, which usually refers to the time point or time range when the image is captured. Then, the AR glasses processor will determine which image or images belong to the target time period based on the preset correspondence (i.e., the correspondence between the image duration and the image), and select the first image from them. The preset correspondence is a key setting that allows the AR glasses processor to accurately find images within the target time period without real-time image acquisition. The preset correspondence may be a database table, an index file or any other form of data structure, which records the acquisition time (or time range) of each image and the associated information of the image itself.

[0043] S130: Analyze the first image to obtain second construction personnel information.

[0044] In this step, the AR glasses processor uses tools such as image processing technology, computer vision algorithms or machine learning models to conduct an in-depth analysis of the received first image. This analysis may include image recognition, feature extraction, object detection and other aspects, aiming to extract useful information from the image. Through the above analysis process, the AR glasses processor can identify key elements in the image, such as construction workers, construction tools, construction environment, etc. Among them, the second construction worker information refers to the specific information related to the construction workers identified and extracted by the AR glasses processor from the image. This information may include the identity of the construction workers (such as name, work number), location, posture, action, etc., depending on the purpose of the analysis and the technical means used.

[0045] In one possible implementation, the first image is analyzed to obtain the second construction worker information, specifically including: using a preset analysis model to extract facial features and identity features from the first image; inputting the facial features and identity features into the preset analysis model respectively to obtain a first recognition result and a second recognition result; obtaining the second construction worker information based on the first recognition result and the second recognition result, the second construction worker information including the construction worker's identity information, position information and authorization information related to the construction task.

[0046] Specifically, the preset analysis model is a pre-trained machine learning model or deep learning model that has the ability to extract useful information from images. This model may be a complex neural network that can accurately identify key information such as facial features and identity identification in images after being trained with a large amount of labeled data. The AR glasses processor uses the preset analysis model to process the first image, first identifying the human face in the image, and extracting facial features from the facial area, such as the shape and position of the eyes, nose, and mouth. At the same time, the AR glasses processor will also try to identify the identity of the construction workers from the image, such as work badges, logos on safety helmets, or identity information directly transmitted through AR glasses, which are all considered identity features. The extracted facial features and identity features are used as input data and sent to different parts or different sub-models of the preset analysis model. These models may have been specially optimized and trained for facial recognition and identity recognition. Through the processing of the preset analysis model, the AR glasses processor will obtain two main recognition results: the first recognition result is based on facial features, which may include information such as the facial similarity score and whether it matches the facial model of a known person; the second recognition result is based on identity features, which may directly give the identity information of the construction worker, such as name, work number, etc. The AR glasses processor will combine the first recognition result and the second recognition result for further logical judgment and data integration. If both point to the same construction worker, then the identity information of the construction worker can be determined with greater certainty. In addition, the AR glasses processor may also further query or calculate the construction worker's position information, authorization information related to the current construction task, etc. based on the construction worker's identity information, and finally obtain the complete second construction worker information.

[0047] S140: If it is determined that the first construction worker information and the second construction worker information are consistent, obtain the second image sent by the AR glasses.

[0048] Specifically, the AR glasses processor compares the first construction worker information and the second construction worker information to determine whether they are consistent. This consistency may refer to the matching of multiple aspects such as identity information, position information, authorization information, etc. If the two are consistent in key information, the AR glasses processor can assume that they represent the same construction worker. Once the consistency of the first construction worker information and the second construction worker information is determined, the AR glasses processor will perform a subsequent action: obtain the second image sent by the AR glasses. This second image may be captured in real time, or it may be a previously stored image but related to the current operation. The purpose of acquiring the second image may be for further analysis, recording, monitoring or other needs related to the construction task.

[0049] For example, suppose that on a construction site, a construction worker named Xiao Zhang is working while wearing AR glasses. The AR glasses processor has previously stored Xiao Zhang's identity information, job information (such as bricklayer), and authorization information (such as permission to enter a specific area). The AR glasses processor stores Xiao Zhang's information, including name, work number, job (bricklayer), authorized area, etc. Through the first image captured by the AR glasses, the AR glasses processor analyzes and confirms that the construction worker in the image is also Xiao Zhang, and extracts his facial features, information on the work badge, etc., and after comparing with the preset model or database, confirms his identity information, job (bricklayer) and authorization information. After comparing the first construction worker information with the second construction worker information, the AR glasses processor finds that the two are consistent in identity information, job information, and authorization information, so it is confirmed that they represent the same construction worker Xiao Zhang. Based on the above consistency judgment, the AR glasses processor sends an instruction to the AR glasses, requiring it to send a second image at the current moment or a specific time point. This image may be used for further analysis (such as checking the quality of Xiao Zhang's work), recording (such as a construction log), or monitoring (such as ensuring that Xiao Zhang works in the authorized area).

[0050] S150, identifying the second image to obtain real-time construction steps.

[0051] Specifically, in this step, the AR glasses processor uses image processing, computer vision or machine learning and other technical means to conduct an in-depth analysis of the received second image. This analysis aims to extract key information related to the construction steps from the image, such as the actions of construction workers, the use of construction tools, changes in the construction environment, etc. Through the above recognition process, the AR glasses processor can identify the current construction steps reflected in the image. This may involve classifying the actions of construction workers to determine which specific construction operation they are performing; or identifying construction tools and materials to determine their role and status in the current construction step. Ultimately, the AR glasses processor will integrate this information into a clear real-time description of the construction steps.

[0052] In one possible implementation, the second image is recognized to obtain real-time construction steps, specifically including: extracting image frames from the second image to obtain multiple image frames; arranging and combining the multiple image frames according to a preset time sequence to obtain a set of images to be recognized; using a preset analysis model to extract behavioral features from multiple image frames in the set of images to be recognized to obtain a behavior feature set; and performing feature comparison between the behavior feature set and a preset behavior feature set contained in a preset construction step corresponding to a preset construction task to obtain real-time construction steps.

[0053] Specifically, the second image may be a frame in a video stream, or a set of multiple static images. In any case, the AR glasses processor first needs to extract multiple image frames from the second image. These image frames represent continuous or critical moments in the construction process. The extracted image frames need to be arranged and combined according to a preset time sequence. This time sequence may be based on the capture time of the image frame, the timestamp in the image content, or other time reference information. The image frame set after arrangement and combination is called the image set to be identified, which reflects the continuity and dynamics of the construction process. The AR glasses processor uses a preset analysis model to extract behavioral features from each image frame in the image set to be identified. This preset analysis model may be based on deep learning, machine learning, or traditional image processing algorithms, which can identify key behavioral features such as the actions of construction workers, the use of construction tools, and changes in the construction environment in the image frame. The extracted behavioral feature set contains detailed information describing the current construction steps. The extracted behavioral feature set needs to be compared with the preset behavioral feature set in the preset construction steps corresponding to the preset construction task. The preset construction steps are defined in advance and include a series of steps required to complete the construction task and the behavioral features that should appear in each step. Through feature comparison, the AR glasses processor can find the preset construction steps that best match the current set of behavioral features, thereby determining the real-time construction steps.

[0054] For example, suppose that on a construction site, a construction worker named Xiao Wang is doing wall masonry work. The AR glasses processor captures the video stream of this process through AR glasses or other devices, and selects one of the time periods as the second image for analysis. The AR glasses processor extracts multiple image frames from the second image (video stream segment), each of which represents a moment of Xiao Wang in the masonry process. These image frames are arranged and combined in the order in which they were captured to form an ordered set of images to be identified. This set shows the entire masonry process of Xiao Wang from picking up bricks, applying mortar to placing bricks on the wall. The AR glasses processor uses a preset analysis model to extract behavioral features from each image frame. For example, in a certain image frame, the model identifies key behavioral features such as Xiao Wang's hand movements (applying mortar), the location of the bricks (in the hand or on the wall), and the state of the mortar (whether it is evenly applied). These features are extracted and form a behavioral feature set. The extracted behavioral feature set is compared with the preset construction steps in the preset construction task "wall masonry". The preset construction steps may include "preparing materials", "applying mortar", "placing bricks", etc., and each step has a corresponding preset behavior feature set. Through comparison, the AR glasses processor finds that the current behavior feature set is the best match with the preset behavior feature set of the step "placing bricks", so it determines that the current real-time construction step is "placing bricks".

[0055] S160. Generate real-time construction prompt information corresponding to the real-time construction steps according to the preset construction step information.

[0056] Specifically, the preset construction step information refers to the information about the standard or recommended construction steps for various construction tasks that are pre-defined in the AR glasses processor. This information may include the specific operation content, required materials, safety precautions, etc. of each step. As before, this is the current construction step that the AR glasses processor identifies from the real-time image of the current construction site through steps such as image recognition, behavioral feature extraction, and feature comparison. Once the real-time construction step is determined, the AR glasses processor will generate corresponding prompt information based on the preset construction step information corresponding to the step. These prompt information may include suggestions for the next operation, precautions for the current step, safety warnings, material preparation prompts, etc., aimed at helping construction workers complete construction tasks more efficiently and safely. Among them, the prompt information is displayed on the screen of the AR glasses.

[0057] In one possible implementation, refer to Figure 2 , Figure 2Another flow chart of a method for processing engineering safety and quality supervision inspection data provided in an embodiment of the present application. According to the preset construction step information, real-time construction prompt information corresponding to the real-time construction step is generated, specifically including steps S210 to S220, the above steps are as follows: S210, extracting the preset construction prompt information corresponding to the preset construction step information from the preset database, the preset database pre-stores the correspondence between the preset construction step information and the preset construction prompt information; S220, marking the preset construction prompt information to the real-time construction step, and obtaining a screen to be displayed, the screen to be displayed includes the real-time construction prompt information.

[0058] Specifically, the preset database is a database that pre-establishes and stores construction-related information. In this database, a clear correspondence is established between the preset construction step information and the preset construction prompt information. This means that for each preset construction step, there are one or more corresponding construction prompt information to guide or remind construction personnel to perform correct operations. When the AR glasses processor determines the current real-time construction step, it will search the preset database for the corresponding preset construction prompt information based on this step information. This process is similar to looking up the meaning of a word in a dictionary. The AR glasses processor will retrieve matching construction prompt information from the database based on the keywords or identifiers of the step information. Once the preset construction prompt information is extracted, the AR glasses processor will annotate this information in some way to the display of the real-time construction steps. This annotation can be in the form of text prompts, icon marks, highlighting, etc., which is intended to attract the attention of construction personnel and clearly inform the supervisor of the next step of the operation points or precautions. The real-time construction steps and their corresponding construction prompt information that have been annotated together constitute the screen to be displayed. This screen is displayed to the construction supervisor through AR glasses.

[0059] In a possible implementation, before using a preset analysis model to extract facial features and identity features from a first image, the preset analysis model is trained; training the preset analysis model specifically includes: obtaining training information; inputting the training information into an adaptive feature fusion network for training to obtain a first training result; superimposing and standardizing the first training result and the training information to obtain a second training result; inputting the second training result into an adaptive feature fusion network for processing to obtain a third training result; superimposing and standardizing the third training result and the second training result until a training information similarity matrix is ​​output, and the training information similarity matrix satisfies a preset logistic regression condition.

[0060] Specifically, before formally using the preset analysis model to extract facial features and identity features from the image, the model needs to be trained first. The purpose of training is to enable the model to accurately identify the required features from the input image. The training information usually includes a large amount of image data and its corresponding labels or annotation information. These image data cover facial images under different angles, lighting conditions, expression changes, etc., as well as corresponding identity information. AFFN is a special neural network structure that can automatically learn and fuse feature information at different levels during the training process. This ability enables AFFN to perform well when processing complex image data. The training information is input into AFFN for initial training to obtain the first training result. This result may not be accurate enough because the model has not fully learned the key features in the image. The first training result is superimposed with the original training information and standardized. The purpose of superposition is to combine the model's preliminary prediction with the true label to provide richer information for the next training; the standardization process is to ensure the consistency and comparability of the data. The processed result is input into AFFN again for training to obtain the third training result. This process is repeated many times, and each time the new training results are superimposed and standardized with the previous results. As the number of iterations increases, the model will gradually learn more accurate and stable feature representations. After multiple iterations of training, AFFN will output a training information similarity matrix. This matrix reflects the similarity relationship between different features in the input image and is one of the important indicators for evaluating model performance. When the training information similarity matrix meets the preset logistic regression conditions, it means that the model training has achieved the expected results. These conditions may include the specific structure of the similarity matrix, stability indicators, accuracy indicators, etc.

[0061] This application also provides a project safety quality supervision inspection data processing system, referring to Figure 3 , Figure 3 A module schematic diagram of a project safety quality supervision inspection data processing system provided in an embodiment of the present application. The data processing system is an AR glasses processor, and the AR glasses processor includes an acquisition module 31 and a processing module 32, wherein the acquisition module 31 acquires the construction plan corresponding to the target time period, and the construction plan includes the first construction personnel information and the preset construction step information; the acquisition module 31 acquires the first image sent by the AR glasses in the target time period, and the user wears the AR glasses; the processing module 32 analyzes the first image to obtain the second construction personnel information; if the processing module 32 determines that the first construction personnel information and the second construction personnel information are consistent, then the processing module 32 acquires the second image sent by the AR glasses; the processing module 32 identifies the second image and obtains the real-time construction steps; the processing module 32 generates the real-time construction prompt information corresponding to the real-time construction steps according to the preset construction step information.

[0062] In a possible implementation, the acquisition module 31 acquires the construction plan corresponding to the target time period, specifically including: the acquisition module 31 receives the plan data uploaded by the user; the processing module 32 preprocesses the plan data to obtain the construction plan corresponding to the target time period, and the preprocessing includes denoising, filtering, correspondence extraction and normalization.

[0063] In a possible implementation, in a target time period, a first image sent by the AR glasses is obtained, and the user wears the AR glasses, specifically including: the processing module 32 determines whether the target time period includes the current moment; if the processing module 32 determines that the target time period includes the current moment, the AR glasses are controlled to capture images in real time to obtain the first image; if the processing module 32 determines that the target time period does not include the current moment, the image duration is obtained from multiple images captured by the AR glasses according to the target time period; the processing module 32 determines the first image based on the image duration and a preset corresponding relationship, and the preset corresponding relationship is the corresponding relationship between the image duration and the image.

[0064] In a possible implementation, the processing module 32 analyzes the first image to obtain the second construction worker information, specifically including: the processing module 32 uses a preset analysis model to extract facial features and identity features from the first image; the processing module 32 inputs the facial features and identity features into the preset analysis model respectively to obtain a first recognition result and a second recognition result; the processing module 32 obtains the second construction worker information based on the first recognition result and the second recognition result, and the second construction worker information includes the construction worker's identity information, position information, and authorization information related to the construction task.

[0065] In a possible implementation, the processing module 32 recognizes the second image and obtains real-time construction steps, which specifically includes: the processing module 32 extracts image frames from the second image to obtain multiple image frames; the processing module 32 arranges and combines the multiple image frames according to a preset time sequence to obtain a set of images to be recognized; the processing module 32 uses a preset analysis model to extract behavioral features from multiple image frames in the set of images to be recognized to obtain a set of behavioral features; the processing module 32 compares the behavioral feature set with a preset behavioral feature set contained in a preset construction step corresponding to a preset construction task to obtain real-time construction steps.

[0066] In a possible implementation, the processing module 32 generates real-time construction prompt information corresponding to the real-time construction steps based on the preset construction step information, specifically including: the processing module 32 extracts the preset construction prompt information corresponding to the preset construction step information from a preset database, and the preset database pre-stores the correspondence between the preset construction step information and the preset construction prompt information; the processing module 32 marks the preset construction prompt information to the real-time construction step to obtain a screen to be displayed, and the screen to be displayed includes the real-time construction prompt information.

[0067] In one possible implementation, the processing module 32 trains the preset analysis model before using the preset analysis model to extract facial features and identity features from the first image; the training of the preset analysis model specifically includes: the acquisition module 31 acquires training information; the processing module 32 inputs the training information into the adaptive feature fusion network for training to obtain a first training result; the processing module 32 superimposes and standardizes the first training result and the training information to obtain a second training result; the processing module 32 inputs the second training result into the adaptive feature fusion network for processing to obtain a third training result; the processing module 32 superimposes and standardizes the third training result and the second training result until a training information similarity matrix is ​​output, and the training information similarity matrix satisfies a preset logistic regression condition.

[0068] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0069] The present application also provides an electronic device, referring to Figure 4 , Figure 4 The electronic device may include: at least one processor 41 , at least one network interface 44 , a user interface 43 , a memory 45 , and at least one communication bus 42 .

[0070] The communication bus 42 is used to realize the connection and communication between these components.

[0071] The user interface 43 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 43 may also include a standard wired interface and a wireless interface.

[0072] The network interface 44 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0073] Among them, the processor 41 may include one or more processing cores. The processor 41 uses various interfaces and lines to connect the various parts of the entire AR glasses processor, and executes various functions and processes data of the AR glasses processor by running or executing instructions, programs, code sets or instruction sets stored in the memory 45, and calling data stored in the memory 45. Optionally, the processor 41 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 41 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 41, but may be implemented separately through a chip.

[0074] Among them, the memory 45 may include a random access memory (RAM) or a read-only memory (Read-Only Memory). Optionally, the memory 45 includes a non-transitory computer-readable storage medium. The memory 45 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 45 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 45 may also be optionally at least one storage device located away from the aforementioned processor 41. As Figure 4 As shown, the memory 45 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program of a method for processing engineering safety quality supervision and inspection data.

[0075] exist Figure 4 In the electronic device shown, the user interface 43 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 41 can be used to call an application program stored in the memory 45 for a method for processing engineering safety and quality supervision inspection data. When executed by one or more processors, the electronic device executes one or more methods in the above-mentioned embodiments.

[0076] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0077] The present application also provides a computer-readable storage medium, which stores instructions. When executed by one or more processors, the electronic device executes one or more of the methods described in the above embodiments.

[0078] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0079] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0080] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0081] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0082] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including several instructions for a computer device (which can be a personal computer, AR glasses processor or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: various media that can store program codes, such as USB flash drives, mobile hard drives, magnetic disks or optical disks.

[0083] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification and the truth of practice, it will be easy for those skilled in the art to think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are regarded as exemplary only, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for processing engineering safety and quality supervision inspection data, characterized in that: The method comprises: Obtaining a construction plan corresponding to a target time period, wherein the construction plan includes first construction personnel information and preset construction step information; In the target time period, acquiring a first image sent by the AR glasses, the user wearing the AR glasses; Analyze the first image to obtain second construction personnel information; If it is determined that the first construction worker information and the second construction worker information are consistent, acquiring a second image sent by the AR glasses; Recognize the second image to obtain real-time construction steps; According to the preset construction step information, real-time construction prompt information corresponding to the real-time construction step is generated.

2. The engineering safety quality supervision inspection data processing method according to claim 1 is characterized in that: The obtaining of the construction plan corresponding to the target time period specifically includes: Receiving the plan data uploaded by the user; The planning data is preprocessed to obtain a construction plan corresponding to the target time period, wherein the preprocessing includes denoising, filtering, corresponding relationship extraction and normalization processing.

3. The engineering safety quality supervision inspection data processing method according to claim 1 is characterized in that: The step of acquiring, in the target time period, a first image sent by the AR glasses, wherein the user wears the AR glasses, specifically includes: Determine whether the target time period includes the current time; If it is determined that the target time period includes the current time, controlling the AR glasses to collect images in real time to obtain the first image; If it is determined that the target time period does not include the current moment, searching for the image duration according to the target time period from multiple images captured by the AR glasses; The first image is determined according to the image duration in combination with a preset corresponding relationship, where the preset corresponding relationship is a corresponding relationship between image duration and images.

4. The engineering safety quality supervision inspection data processing method according to claim 1 is characterized in that: The analyzing the first image to obtain the second construction personnel information specifically includes: Using a preset analysis model, extracting facial features and identity features from the first image; Inputting the facial features and the identity features into the preset analysis model respectively to obtain a first recognition result and a second recognition result; The second construction worker information is obtained according to the first recognition result and the second recognition result, where the second construction worker information includes the construction worker's identity information, position information, and authorization information related to the construction task.

5. The engineering safety quality supervision inspection data processing method according to claim 1 is characterized in that: The identifying the second image to obtain real-time construction steps specifically includes: Extracting image frames from the second image to obtain a plurality of image frames; Arrange and combine the plurality of image frames according to a preset time sequence to obtain a set of images to be identified; Using a preset analysis model to extract behavior features from a plurality of image frames in the image set to be identified, to obtain a behavior feature set; The behavior feature set is compared with a preset behavior feature set included in a preset construction step corresponding to a preset construction task to obtain the real-time construction step.

6. The engineering safety quality supervision inspection data processing method according to claim 1 is characterized in that: Generating the real-time construction prompt information corresponding to the real-time construction step according to the preset construction step information specifically includes: Extracting preset construction prompt information corresponding to the preset construction step information from a preset database, wherein the preset database pre-stores a corresponding relationship between the preset construction step information and the preset construction prompt information; The preset construction prompt information is marked in the real-time construction step to obtain a picture to be displayed, and the picture to be displayed includes the real-time construction prompt information.

7. The method for processing engineering safety quality supervision inspection data according to claim 4 is characterized in that: Before extracting facial features and identity features from the first image using the preset analysis model, training the preset analysis model; the training of the preset analysis model specifically includes: Get training information; Inputting the training information into an adaptive feature fusion network for training to obtain a first training result; After superimposing and normalizing the first training result and the training information, a second training result is obtained; Inputting the second training result into the adaptive feature fusion network for processing to obtain a third training result; The third training result and the second training result are superimposed and standardized until the training information similarity matrix is ​​output, and the training information similarity matrix satisfies a preset logistic regression condition.

8. A project safety quality supervision inspection data processing system, characterized in that: The data processing system comprises an acquisition module (31) and a processing module (32), wherein: The acquisition module (31) is used to acquire a construction plan corresponding to the target time period, wherein the construction plan includes first construction personnel information and preset construction step information; The acquisition module (31) is further used to acquire a first image sent by the AR glasses in the target time period, and the user is wearing the AR glasses; The processing module (32) is used to analyze the first image to obtain second construction personnel information; The processing module (32) is further configured to obtain a second image sent by the AR glasses if it is determined that the first construction worker information and the second construction worker information are consistent; The processing module (32) is further used to identify the second image and obtain real-time construction steps; The processing module (32) is also used to generate real-time construction prompt information corresponding to the real-time construction step according to the preset construction step information.

9. An electronic device, characterized in that: The electronic device comprises a processor (41), a memory (45), a user interface (43) and a network interface (44), wherein the memory (45) is used to store instructions, the user interface (43) and the network interface (44) are both used to communicate with other devices, and the processor (41) is used to execute the instructions stored in the memory (45) so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images