Construction site information verification processing method and system based on image analysis

Abstract

The embodiment of the present application provides a construction site information verification and processing method and system based on image analysis, which relates to the field of artificial intelligence technology, and can accurately capture the image data of the real construction behavior of the construction site nodes and their derived collaborative behavior, which not only improves the comprehensiveness and accuracy of data collection, but also significantly enhances the real-time and dynamic adaptability of construction site monitoring. By introducing node scene feature analysis, it is possible to accurately identify and quantify monitoring noise errors, and then optimize the original image data, significantly reducing the impact of external interference factors on the monitoring results, ensuring the reliability and accuracy of monitoring data, and greatly improving the quality of monitoring data. Finally, by comparing and analyzing the optimized construction behavior and collaborative behavior image data of the nodes in the test cluster and the reference cluster, it is possible to scientifically and objectively evaluate the effectiveness of different monitoring schemes, providing strong data support and scientific basis for construction site management decisions.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a construction site information verification and processing method and system based on image analysis. Background Art

[0002] In the process of modern building construction, monitoring and management of construction sites are important links to ensure project quality and progress. Traditional methods of construction site monitoring often rely on manual inspections and paper records, which are not only inefficient but also prone to errors and difficult to accurately reflect the actual situation of the construction site in a timely manner. With the rapid development of information technology, especially the mature application of image processing and analysis technology, new solutions have been provided for construction site monitoring.

[0003] However, although the current image monitoring technology can realize remote visual monitoring of construction sites, it still has many deficiencies in information processing and analysis. On the one hand, the construction site environment is complex and changeable. Factors such as lighting conditions, weather conditions, and dynamic changes in construction machinery and personnel may affect the quality of monitoring images, resulting in a large amount of noise and interference information in the image data, affecting the accuracy of the monitoring results. On the other hand, the collaborative operation characteristics of the construction site make the various construction nodes closely interconnected. The construction behavior of a single node often affects other nodes, and the existing monitoring system often lacks the ability to effectively capture and analyze such derivative monitoring execution data.

[0004] In addition, the application effects of different monitoring schemes on construction sites vary significantly. How to scientifically evaluate and optimize monitoring schemes to improve monitoring efficiency and quality has become an urgent problem to be solved. At present, there is a lack of a systematic method on the market to fully verify the effectiveness of monitoring schemes. Especially in complex construction environments with multiple nodes and multiple scenarios, how to ensure the authenticity and reliability of monitoring data, and how to compare and analyze the effects of different monitoring schemes have become research hotspots in the field of construction site information management. Summary of the invention

[0005] In order to at least overcome the above-mentioned deficiencies in the prior art, the purpose of the present application is to provide a construction site information verification and processing method and system based on image analysis.

[0006] In a first aspect, the present application provides a construction site information verification processing method based on image analysis, which is applied to a construction site information verification processing system based on image analysis, and the method comprises:

[0007] Add X construction site nodes to a test cluster and a reference cluster, configure first monitoring execution data to the construction site nodes in the test cluster according to a test monitoring scheme, and configure second monitoring execution data to the construction site nodes in the reference cluster according to a reference monitoring scheme; X is an integer greater than 1;

[0008] Acquire real construction behavior image data of each construction site node among the X construction site nodes for the configured monitoring execution data, and acquire real construction collaborative behavior image data of the derived monitoring execution data derived and extended from each construction site node; the configured monitoring execution data is the monitoring execution data sent to each construction site node after configuration, and both the configured monitoring execution data and the derived monitoring execution data belong to the time series formed by the first monitoring execution data and the second monitoring execution data;

[0009] Determine the node scene feature of each construction site node for the monitoring noise error corresponding to the construction behavior event of the monitoring execution data of each construction site node, optimize the real construction behavior image data and the real construction collaborative behavior image data corresponding to the X construction site nodes respectively based on the monitoring noise errors corresponding to the X construction site nodes respectively, and generate the optimized construction behavior image data and the optimized construction collaborative behavior image data corresponding to the X construction site nodes respectively;

[0010] Based on the optimized construction behavior image data of the construction site nodes in the test cluster and the construction behavior state error information between the optimized construction behavior image data of the construction site nodes in the reference cluster, as well as the optimized construction collaborative behavior image data of the construction site nodes in the test cluster and the collaborative behavior state error information between the optimized construction collaborative behavior image data of the construction site nodes in the reference cluster, the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme are determined respectively.

[0011] In a second aspect, an embodiment of the present application further provides a construction site information verification and processing system based on image analysis, the construction site information verification and processing system based on image analysis comprising a processor and a machine-readable storage medium, the machine-readable storage medium storing a computer program, the computer program being loaded and executed in conjunction with the processor to implement the construction site information verification and processing method based on image analysis of the above first aspect.

[0012] In combination with the above aspects, the embodiment of the present application effectively realizes the multi-dimensional monitoring and comparative analysis of construction site behavior by constructing a test cluster and a reference cluster, and respectively configuring different monitoring execution data. The method can accurately capture the image data of the real construction behavior of the construction site nodes and their derived collaborative behavior, which not only improves the comprehensiveness and accuracy of data collection, but also significantly enhances the real-time and dynamic adaptability of construction site monitoring. By introducing node scene feature analysis, the method can accurately identify and quantify monitoring noise errors, and then optimize the original image data, significantly reducing the impact of external interference factors on the monitoring results, and ensuring the reliability and accuracy of monitoring data. This optimization process greatly improves the quality of monitoring data and provides a solid foundation for subsequent analysis. Finally, by comparing and analyzing the optimized construction behavior and collaborative behavior image data of the nodes in the test cluster and the reference cluster, the effectiveness of different monitoring schemes can be scientifically and objectively evaluated, providing strong data support and scientific basis for construction site management decisions. It not only helps to optimize the allocation of monitoring resources and improve monitoring efficiency, but also promotes the refinement and intelligence level of construction site management, and significantly improves project execution efficiency and engineering quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 A schematic diagram of a process flow of a construction site information verification and processing method based on image analysis provided in an embodiment of the present application;

[0015] Figure 2 A structural schematic block diagram of a construction site information verification and processing system based on image analysis for implementing the above-mentioned construction site information verification and processing method based on image analysis provided in an embodiment of the present application. DETAILED DESCRIPTION

[0016] The following description is intended to enable one of ordinary skill in the art to implement and utilize the present application, and the description is provided in the context of a specific application scenario and its requirements. It will be apparent to one of ordinary skill in the art that various changes may be made to the disclosed embodiments, and that the general principles defined in the present application may be applied to other embodiments and application scenarios without departing from the principles and scope of the present application. Therefore, the present application is not limited to the described embodiments, but should be given the broadest scope equal to the claims.

[0017] Step S110, adding X construction site nodes to the test cluster and the reference cluster, configuring first monitoring execution data to the construction site nodes in the test cluster according to the test monitoring scheme, and configuring second monitoring execution data to the construction site nodes in the reference cluster according to the reference monitoring scheme. X is an integer greater than 1.

[0018] In detail, the construction site node refers to a specific unit or area in the construction site, which can be monitored and managed independently. For example, a large construction site can be divided into multiple construction areas, each of which can be regarded as a construction site node. These construction site nodes may include different construction stages, different construction tasks or different construction teams.

[0019] The test cluster is a collection of nodes on the construction site used to test a new or modified monitoring scheme. These nodes will execute according to the new monitoring scheme to evaluate its effectiveness.

[0020] The reference cluster is a set of nodes on the construction site serving as a control group, which are usually executed according to an existing or known monitoring scheme so as to be compared with the test cluster and to evaluate the advantages or disadvantages of the new scheme.

[0021] The monitoring execution data refers to the specific instructions or requirements assigned to the construction site nodes in order to implement the monitoring plan. These monitoring execution data may include the type of information to be collected (such as construction progress, construction quality, construction safety, etc.), collection frequency, report format, etc.

[0022] Thus, the first monitoring execution data is monitoring execution data configured for the construction site nodes in the test cluster according to the test monitoring scheme. The second monitoring execution data is monitoring execution data configured for the construction site nodes in the reference cluster according to the reference monitoring scheme.

[0023] Exemplarily, in this embodiment, it is assumed that there are five construction site nodes, which are respectively marked as node A, node B, node C, node D and node E. These five construction site nodes are added to the test cluster and the reference cluster.

[0024] The test monitoring scheme aims to focus on monitoring the construction progress and quality of the construction site. According to this scheme, the first monitoring execution data configured to the construction site nodes (such as node A, node B and node C) in the test cluster may include: requiring node A to upload construction progress photos every 2 hours and record the completion time and quality evaluation of each process in detail; requiring node B to submit the use and quality inspection report of construction materials every day; requiring node C to monitor the operating status and fault information of construction equipment in real time.

[0025] The reference monitoring scheme focuses on monitoring construction safety and personnel management. According to this scheme, the second monitoring execution data configured for the construction site nodes (such as nodes D and E) in the reference cluster may be: requiring node D to conduct a safety inspection of the construction area every 4 hours and record the safety hazards found; requiring node E to count the attendance and working hours of construction personnel in real time.

[0026] When the server performs this step, it will accurately send the corresponding monitoring execution data to each construction site node through the network, and ensure that the nodes can receive and understand these instructions.

[0027] Step S120, obtaining real construction behavior image data of each of the X construction site nodes for the configuration monitoring execution data, and obtaining real construction collaborative behavior image data of the derived monitoring execution data derived and extended by each of the construction site nodes. The configuration monitoring execution data is the monitoring execution data sent to each of the construction site nodes after configuration, and both the configuration monitoring execution data and the derived monitoring execution data belong to the time series formed by the first monitoring execution data and the second monitoring execution data.

[0028] In detail, the real construction behavior image data refers to the graphical representation of the actual construction behavior performed by the construction site nodes according to the configured monitoring execution data. These real construction behavior image data may include photos, videos, charts, etc., which are used to intuitively display the construction progress, quality, safety, etc.

[0029] The derived monitoring execution data refers to the corresponding monitoring execution data of other construction site nodes caused by the change of construction behavior of one construction site node. For example, the early construction progress of one node may cause other nodes to adjust the construction plan or resource allocation. The real construction collaborative behavior image data refers to the graphical representation of the collaborative construction behavior generated due to the mutual influence between the construction site nodes, reflecting the construction collaboration between different nodes, such as resource allocation, progress coordination, etc.

[0030] In this embodiment, the server starts to obtain the real construction behavior image data of the five construction site nodes for the configured monitoring execution data, as well as the real construction collaborative behavior image data of the derived monitoring execution data derived and expanded from each construction site node.

[0031] Taking node A as an example, it uploads construction progress photos every 2 hours according to the requirements of the test monitoring plan. These photos are the real construction behavior image data for the configuration monitoring execution data. At the same time, the change of node A's construction progress may affect the construction arrangement of node B. For example, if the progress of node A is ahead of schedule, node B may speed up the construction pace. The construction behavior of node B caused by the change of node A is the real construction collaborative behavior of the derived monitoring execution data derived and expanded from node A, and its related image data is the real construction collaborative behavior image data.

[0032] For node D, a safety inspection of the construction area is carried out every 4 hours according to the reference monitoring plan, and the safety hazards found are recorded. These inspection records and photos of hidden dangers are the real construction behavior image data. Since some safety hazards found by node D may cause node E to adjust the work arrangements of personnel, this adjustment behavior of node E is the real construction collaborative behavior derived and extended from node D, and the relevant image data is the real construction collaborative behavior image data.

[0033] The server receives and stores these real construction behavior image data and real construction collaborative behavior image data in real time through the communication connection established with each construction site node for subsequent processing and analysis.

[0034] Step S130, determining the monitoring noise error corresponding to the construction behavior event of the monitoring execution data of each construction site node for the node scene feature of each construction site node, optimizing the real construction behavior image data and the real construction collaborative behavior image data corresponding to the X construction site nodes respectively based on the monitoring noise errors corresponding to the X construction site nodes respectively, and generating the optimized construction behavior image data and the optimized construction collaborative behavior image data corresponding to the X construction site nodes respectively.

[0035] In detail, the node scene characteristics refer to the external environment and internal conditions that affect the construction behavior of the nodes at the construction site. These factors may include weather conditions, topography, construction equipment performance, construction personnel skill level, etc. The monitoring noise error refers to the error in the monitoring results caused by factors other than the monitoring execution data itself, such as node scene characteristics. For example, bad weather may cause delays in the construction progress, which is not directly caused by the monitoring execution data, but is affected by the external environment.

[0036] The optimized construction behavior image data and optimized construction collaborative behavior image data refer to image data obtained by performing noise correction and optimization processing on the real construction behavior image data and the real construction collaborative behavior image data. These optimized data more accurately reflect the actual situation of the construction site and reduce the impact of monitoring noise errors.

[0037] In this embodiment, the server determines the monitoring noise error corresponding to the construction behavior event of each construction site node with respect to the monitoring execution data of each construction site node, based on the node scene feature of each construction site node.

[0038] Node scenario characteristics can include environmental conditions at the construction site (such as weather and terrain), the performance of construction equipment, the skill level of construction personnel, etc.

[0039] Take node A as an example. Assume that the weather at the construction site is bad, which affects the construction progress. This weather factor is considered as monitoring noise. The server estimates the noise error weight of node A due to weather factors for construction behavior events (such as completing a specific process on time) through the noise error estimation network.

[0040] Then, the server calculates the mean of the noise error weights corresponding to the five construction site nodes and generates a benchmark noise error weight.

[0041] Assuming that the noise error weight of node A is significantly higher than the baseline noise error weight, the server obtains the loss value between the two and determines this loss value as the monitoring noise error of node A.

[0042] Based on the monitoring noise errors corresponding to the five construction site nodes, the server optimizes the real construction behavior image data and the real construction collaborative behavior image data.

[0043] For node A, the server first calculates the mean of the real construction behavior image data of the other four nodes (node ​​B, node C, node D and node E) to generate benchmark construction behavior image data; and calculates the mean of the real construction collaborative behavior image data of the other four nodes to generate benchmark construction collaborative behavior image data.

[0044] Next, the server obtains the first image feature difference between the real construction behavior image data corresponding to node A and the benchmark construction behavior image data, and the second image feature difference between the real construction collaborative behavior image data corresponding to node A and the benchmark construction collaborative behavior image data.

[0045] According to the noise correction coefficient corresponding to node A, the first image feature difference is subjected to noise correction to generate a first corrected image feature difference; and the second image feature difference is subjected to noise correction to generate a second corrected image feature difference.

[0046] Finally, the server fuses the first corrected image feature difference with the benchmark construction behavior image data to generate the optimized construction behavior image data corresponding to node A; and fuses the second corrected image feature difference with the benchmark construction collaborative behavior image data to generate the optimized construction collaborative behavior image data corresponding to node A.

[0047] The same process is also applied to other construction site nodes to generate their corresponding optimized construction behavior image data and optimized construction collaborative behavior image data.

[0048] Step S140, based on the construction behavior state error information between the optimized construction behavior image data of the construction site nodes in the test cluster and the optimized construction behavior image data of the construction site nodes in the reference cluster, and the collaborative behavior state error information between the optimized construction collaborative behavior image data of the construction site nodes in the test cluster and the optimized construction collaborative behavior image data of the construction site nodes in the reference cluster, determine the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme respectively.

[0049] In detail, the construction behavior state error information is the difference information in the construction behavior between the construction site nodes in the test cluster and the reference cluster, which is used to evaluate the effectiveness of the test monitoring scheme in construction behavior monitoring. The collaborative behavior state error information is the difference information in the construction collaborative behavior between the construction site nodes in the test cluster and the reference cluster, which is used to evaluate the effectiveness of the test monitoring scheme in construction collaborative monitoring.

[0050] The monitoring effectiveness refers to the evaluation results of the effect of the monitoring plan in actual application. An effective monitoring plan can accurately and timely reflect the actual situation of the construction site and help managers make correct decisions and adjustments.

[0051] In this embodiment, the server determines the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme based on the construction behavior state error information between the optimized construction behavior image data of the construction site nodes (node ​​A, node B, and node C) in the test cluster and the optimized construction behavior image data of the construction site nodes (node ​​D and node E) in the reference cluster, and the collaborative behavior state error information between the optimized construction collaborative behavior image data of the construction site nodes in the test cluster and the optimized construction collaborative behavior image data of the construction site nodes in the reference cluster.

[0052] First, the server fuses the same optimized construction behavior characteristic parameters (such as the recording method of construction progress) in the optimized construction behavior image data corresponding to the construction site nodes in the test cluster, and generates the construction behavior characteristic global parameters corresponding to the optimized construction behavior characteristic parameters of the test cluster. Similarly, the server fuses the same optimized construction collaborative behavior characteristic parameters in the optimized construction collaborative behavior image data corresponding to the construction site nodes in the test cluster, and generates the optimized construction collaborative behavior characteristic global parameters corresponding to the optimized construction collaborative behavior characteristic parameters of the test cluster.

[0053] Then, the server obtains the construction behavior feature global parameters corresponding to the corresponding optimized construction behavior feature parameters of the reference cluster, and the optimized construction collaborative behavior feature global parameters corresponding to the optimized construction collaborative behavior feature parameters.

[0054] Next, the server performs loss calculation on the global parameters of the construction behavior characteristics corresponding to the same optimized construction behavior characteristic parameters in the test cluster and the reference cluster, and generates the first loss calculation result corresponding to the optimized construction behavior characteristic parameters. For example, for the characteristic parameter of the recording method of the construction progress, the corresponding first loss calculation result is calculated. Such calculation is performed on all optimized construction behavior characteristic parameters to obtain multiple first loss calculation results, which are output as construction behavior state error information.

[0055] Similarly, the server performs loss calculation on the global parameters of the optimized construction collaborative behavior characteristics corresponding to the same optimized construction collaborative behavior characteristic parameters in the test cluster and the reference cluster, and generates a second loss calculation result corresponding to the optimized construction collaborative behavior characteristic parameters. All optimized construction collaborative behavior characteristic parameters are calculated to obtain multiple second loss calculation results, which are output as collaborative behavior state error information.

[0056] Based on these first loss calculation results and second loss calculation results, the server performs weighted calculation on the associated first loss calculation results and second loss calculation results to generate a plurality of first solution metrics for the construction behavior of the test monitoring solution and the reference monitoring solution.

[0057] For example, if the first loss calculation result of the construction progress is associated with the second loss calculation result of the personnel management, the server will perform weighted calculation on them to obtain a first solution metric value.

[0058] Then, the server obtains the number of the first scheme metric values ​​that are greater than the first threshold value. Assume that the first threshold value is 0.8 and the second threshold value is 3. If the number of the first scheme metric values ​​that are greater than 0.8 is not less than 3, it is determined that the monitoring effectiveness corresponding to the test monitoring scheme is better than the monitoring effectiveness corresponding to the reference monitoring scheme; if the number is less than 3, it is determined that the monitoring effectiveness corresponding to the reference monitoring scheme is better than the monitoring effectiveness corresponding to the test monitoring scheme.

[0059] In addition, the server can also add more construction site nodes (for example, 8 nodes, respectively marked as node F, node G, node H, node I, node J, node K, node L and node M) to the verification cluster and the comparison cluster. According to the test monitoring plan, the third monitoring execution data is configured for the construction site nodes (such as node F, node G and node H) in the verification cluster, and according to the reference monitoring plan, the fourth monitoring execution data is configured for the construction site nodes (such as node I, node J, node K, node L and node M) in the comparison cluster.

[0060] Based on the optimized construction behavior image data of the construction site nodes within the verification cluster, the optimized construction behavior image data of the construction site nodes within the comparison cluster, the optimized construction collaborative behavior image data of the construction site nodes within the verification cluster, and the optimized construction collaborative behavior image data of the construction site nodes within the comparison cluster, multiple second scheme measurement values ​​for the operation of the test monitoring scheme and the reference monitoring scheme are determined.

[0061] Finally, the server comprehensively determines the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme respectively based on the above-mentioned first scheme metric value and the second scheme metric value.

[0062] In addition, the server can also divide these nodes based on the connection topology between multiple construction site nodes (for example, 20 nodes) to generate multiple node groups. Assuming that the connection topology shows the communication relationship between the nodes, four node groups are generated based on the close communication between the nodes and the relevance of the construction tasks. From these four node groups, two node groups that meet the test standards (such as construction scale, construction type, etc.) are extracted, and the construction site nodes in these two node groups are used as the X construction site nodes used in the above steps for the subsequent monitoring scheme effectiveness evaluation.

[0063] Based on the above steps, the embodiment of the present application effectively realizes the multi-dimensional monitoring and comparative analysis of the construction site behavior by constructing a test cluster and a reference cluster, and respectively configuring different monitoring execution data. The method can accurately capture the image data of the real construction behavior and its derived collaborative behavior of the construction site nodes, which not only improves the comprehensiveness and accuracy of data collection, but also significantly enhances the real-time and dynamic adaptability of the construction site monitoring. By introducing node scene feature analysis, the method can accurately identify and quantify the monitoring noise error, and then optimize the original image data, significantly reducing the impact of external interference factors on the monitoring results, and ensuring the reliability and accuracy of the monitoring data. This optimization process greatly improves the quality of the monitoring data and provides a solid foundation for subsequent analysis. Finally, by comparing and analyzing the optimized construction behavior and collaborative behavior image data of the nodes in the test cluster and the reference cluster, the effectiveness of different monitoring schemes can be scientifically and objectively evaluated, providing strong data support and scientific basis for construction site management decisions. It not only helps to optimize the allocation of monitoring resources and improve monitoring efficiency, but also promotes the refinement and intelligence level of construction site management, and significantly improves project execution efficiency and engineering quality.

[0064] In a possible implementation, step S130 includes:

[0065] Step S131: Based on the monitoring noise error corresponding to the ath construction site node among the X construction site nodes, determine the noise correction coefficient corresponding to the ath construction site node. a is a positive integer not greater than X.

[0066] Step S132, performing mean calculation on the real construction behavior image data of the remaining construction site nodes to generate benchmark construction behavior image data, performing mean calculation on the real construction collaborative behavior image data of the remaining construction site nodes to generate benchmark construction collaborative behavior image data. The remaining construction site nodes are the construction site nodes in the cluster where the a-th construction site node is located except the a-th construction site node.

[0067] Step S133, based on the noise correction coefficient corresponding to the a-th construction site node, the benchmark construction behavior image data and the benchmark construction collaborative behavior image data, the real construction behavior image data and the real construction collaborative behavior image data corresponding to the a-th construction site node are optimized to generate the optimized construction behavior image data and the optimized construction collaborative behavior image data corresponding to the a-th construction site node.

[0068] In this embodiment, the following is a very detailed scenario example of the steps in the above technical content:

[0069] Still taking the five construction site nodes mentioned previously (node ​​A, node B, node C, node D and node E) as an example.

[0070] First, the monitoring noise error corresponding to each node is determined and the noise correction coefficient is calculated.

[0071] For node A, assuming that its monitoring noise error is due to construction delays caused by bad weather, a specific algorithm and model is used to calculate the monitoring noise error value corresponding to node A to be 0.3. Based on this error value, the noise correction coefficient corresponding to node A is further determined to be 0.8 through a preset function or rule.

[0072] Next, the reference construction activity image data and the reference construction coordination activity image data are calculated.

[0073] In the test cluster (node ​​A, node B, and node C) where node A is located, node A is removed and the mean of the real construction behavior image data of nodes B and C is calculated. Assume that the values ​​of the real construction behavior image data of node B on a certain feature dimension are 10, 20, and 30, respectively, and the values ​​of the real construction behavior image data of node C on the same feature dimension are 15, 25, and 35, respectively. Then, on this feature dimension, the values ​​of the benchmark construction behavior image data are (10+15) / 2=12.5, (20+25) / 2=22.5, and (30+35) / 2=32.5.

[0074] Similarly, the actual construction collaborative behavior image data of nodes B and C are averaged to obtain the benchmark construction collaborative behavior image data.

[0075] Then, based on the noise correction coefficient, the benchmark construction behavior image data, and the benchmark construction collaborative behavior image data corresponding to node A, the real construction behavior image data and the real construction collaborative behavior image data of node A are optimized.

[0076] Assume that the values ​​of the real construction behavior image data of node A in the above feature dimensions are 8, 18, and 28, respectively, which are different from the values ​​of the benchmark construction behavior image data in this dimension, 12.5, 22.5, and 32.5. Calculate the first image feature difference, which is -4.5, -4.5, and -4.5 in this dimension.

[0077] According to the noise correction coefficient 0.8 corresponding to node A, noise correction is performed on the first image feature difference, and the corrected first image feature differences are -3.6, -3.6, and -3.6 in this dimension, respectively.

[0078] The corrected first image feature difference is fused with the benchmark construction behavior image data. In this dimension, the fusion results are 12.5-3.6=8.9, 22.5-3.6=18.9, and 32.5-3.6=28.9, respectively, thereby generating the optimized construction behavior image data corresponding to node A.

[0079] For the real construction collaborative behavior image data of node A, the second image feature difference is calculated in the same way, noise correction is performed, and it is fused with the benchmark construction collaborative behavior image data to generate the optimized construction collaborative behavior image data corresponding to node A.

[0080] The same process is followed for node B. Assume that the monitoring noise error calculation result of node B is 0.2, and the noise correction coefficient is determined to be 0.9.

[0081] When calculating the benchmark construction behavior image data and the benchmark construction collaborative behavior image data, the real construction behavior image data and the real construction collaborative behavior image data of the node A and the node C are averaged.

[0082] Then, based on the noise correction coefficient of node B, the calculated benchmark data, and the original real construction behavior image data and real construction collaborative behavior image data of node B, difference calculation, noise correction and fusion processing are performed to finally generate the optimized construction behavior image data and optimized construction collaborative behavior image data corresponding to node B.

[0083] For node C, node D and node E, the above steps and methods are also followed to complete the optimization processing in turn, and the corresponding optimized construction behavior image data and optimized construction collaborative behavior image data are obtained.

[0084] Throughout the entire process, the server needs to efficiently process a large amount of data calculation and fusion operations to ensure that the optimization results of each node are accurate and reliable. At the same time, the server must continuously monitor the data processing process to promptly discover and handle possible abnormal situations, such as missing data and calculation errors. Through such fine processing and optimization, it can provide a more accurate and valuable data foundation for the subsequent evaluation of the effectiveness of the test monitoring plan and the reference monitoring plan.

[0085] During the calculation and optimization process, the server will make full use of its powerful computing power and storage resources to ensure the speed and accuracy of data processing. The intermediate and final results of the calculation process will be properly stored and managed for subsequent query and analysis.

[0086] In addition, the server will dynamically adjust the calculation and optimization strategies according to the actual situation. For example, if the data of some nodes is found to be abnormal or not in line with expectations, the data collection and processing process will be rechecked, or the calculation method of the noise correction coefficient will be adjusted to ensure that the optimized construction behavior image data and optimized construction collaborative behavior image data generated in the end can truly reflect the situation on the construction site, providing strong support for the evaluation of the monitoring plan.

[0087] During the optimization process, the server also considers the correlation and mutual influence between different nodes. For example, if the change in the construction behavior of node A has a significant impact on nodes B and C, then when calculating and optimizing the data of nodes B and C, this influence factor will be fully considered to more accurately evaluate the status of the entire construction site and the effectiveness of the monitoring plan.

[0088] Through detailed processing and optimization of these five construction site nodes, the server successfully prepared high-quality data for the subsequent effectiveness evaluation of the monitoring plan, so that it can more accurately judge the advantages and disadvantages of the test monitoring plan and the reference monitoring plan in construction site management, and provide a strong basis for further improvement and optimization of the monitoring plan.

[0089] In a possible implementation, step S133 includes:

[0090] Step S1331, obtaining the first image feature difference between the real construction behavior image data corresponding to the a-th construction site node and the benchmark construction behavior image data, and the second image feature difference between the real construction collaborative behavior image data corresponding to the a-th construction site node and the benchmark construction collaborative behavior image data.

[0091] Step S1332: Perform noise correction on the first image feature difference according to the noise correction coefficient corresponding to the a-th construction site node to generate a first corrected image feature difference; perform noise correction on the second image feature difference according to the noise correction coefficient corresponding to the a-th construction site node to generate a second corrected image feature difference.

[0092] Step S1333: fuse the first corrected image feature difference with the benchmark construction behavior image data to generate the optimized construction behavior image data corresponding to the a-th construction site node; fuse the second corrected image feature difference with the benchmark construction collaborative behavior image data to generate the optimized construction collaborative behavior image data corresponding to the a-th construction site node.

[0093] In this embodiment, the five construction site nodes (node ​​A, node B, node C, node D, and node E) mentioned above are still taken as an example to explain in detail the optimization process of node A.

[0094] First, obtain the first image feature difference between the real construction behavior image data and the benchmark construction behavior image data corresponding to node A. Assuming that in the feature dimension of construction progress, the progress represented by the real construction behavior image data of node A is 80%, while the progress represented by the benchmark construction behavior image data is 70%, then the first image feature difference in this dimension is 10%.

[0095] At the same time, obtain the second image feature difference between the real construction collaborative behavior image data and the benchmark construction collaborative behavior image data corresponding to node A. Assuming that in the dimension of construction personnel collaborative efficiency, the efficiency represented by the real construction collaborative behavior image data of node A is 75%, and the efficiency represented by the benchmark construction collaborative behavior image data is 65%, then the second image feature difference in this dimension is 10%.

[0096] Next, the first image feature difference is subjected to noise correction according to the noise correction coefficient corresponding to node A. Assuming that the noise correction coefficient of node A is 0.8, the first image feature difference after correction is 10%*0.8=8% in the construction progress dimension.

[0097] Similarly, the noise correction is performed on the feature difference of the second image according to the noise correction coefficient corresponding to node A. The corrected feature difference of the second image is 10%*0.8=8% in the construction personnel collaboration efficiency dimension.

[0098] Then, the first corrected image feature difference is fused with the benchmark construction behavior image data to generate the optimized construction behavior image data corresponding to node A. In terms of construction progress, the fused result is 70%+8%=78%.

[0099] The feature difference of the second corrected image is fused with the benchmark construction collaborative behavior image data to generate the optimized construction collaborative behavior image data corresponding to node A. In terms of the collaborative efficiency of construction personnel, the fused result is 65%+8%=73%.

[0100] The same process is also applied to node B. Assuming that in the feature dimension of material usage, the real construction behavior image data of node B is 50 units, and the benchmark construction behavior image data is 45 units, the first image feature difference is 5 units. The noise correction coefficient of node B is 0.7, and the first image feature difference after correction is 5*0.7=3.5 units. The optimized construction behavior image data after fusion is 45+3.5=48.5 units in the dimension of material usage.

[0101] In the collaborative behavior dimension of construction equipment collaborative operation, assuming that the degree of collaboration represented by the real construction collaborative behavior image data of node B is 80, and the degree of collaboration represented by the benchmark construction collaborative behavior image data is 70, the second image feature difference is 10. The corrected second image feature difference is 10*0.7=7. The fused optimized construction collaborative behavior image data is 70+7=77 in the construction equipment collaborative operation dimension.

[0102] For node C, assume that in the construction quality assessment dimension, its real construction behavior image data is 90 points, the benchmark construction behavior image data is 85 points, and the first image feature difference is 5 points. The noise correction coefficient is 0.9, and the first image feature difference after correction is 5*0.9=4.5 points. The fused optimized construction behavior image data is 85+4.5=89.5 points in the construction quality assessment dimension.

[0103] In the collaborative behavior dimension of construction resource allocation collaboration, it is assumed that the collaborative effect represented by the real construction collaborative behavior image data of node C is 85, the collaborative effect represented by the benchmark construction collaborative behavior image data is 75, and the second image feature difference is 10. The corrected second image feature difference is 10*0.9=9. The fused optimized construction collaborative behavior image data is 75+9=84 in the construction resource allocation collaboration dimension.

[0104] Nodes D and E are also processed according to the same logic and steps. For example, in the behavioral dimension of the number of potential safety hazards, the number of real construction behavior image data for node D is 3, the number of baseline construction behavior image data is 2, and the difference in the first image feature is 1. The noise correction coefficient is assumed to be 0.6, and the difference in the first image feature after correction is 1*0.6=0.6. The optimized construction behavior image data after fusion is 2+0.6=2.6 in the dimension of the number of potential safety hazards.

[0105] In the collaborative behavior dimension of personnel deployment and coordination, it is assumed that the collaborative effect represented by the real construction collaborative behavior image data of node D is 70, the collaborative effect represented by the benchmark construction collaborative behavior image data is 60, and the second image feature difference is 10. The corrected second image feature difference is 10*0.6=6. The fused optimized construction collaborative behavior image data is 60+6=66 in the dimension of personnel deployment and coordination.

[0106] During the entire processing process, the server needs to accurately calculate and process data differences, corrections, and fusion operations in each dimension. This requires not only powerful computing capabilities, but also accurate reading and storage of information such as the original data, benchmark data, and correction coefficients of each node.

[0107] During the calculation process, the server must ensure the consistency and accuracy of the data to avoid data loss or miscalculation. For each feature dimension of each node, it must be processed strictly according to the predetermined algorithm and process to generate reliable optimized construction behavior image data and optimized construction collaborative behavior image data.

[0108] At the same time, the server will monitor the progress and results of data processing in real time. If there is an abnormality in the processing results of a certain node or a certain feature dimension, the server will immediately issue an alarm and conduct a backtracking check to find out the problem and correct it.

[0109] By processing and optimizing each node, the server can provide high-quality, high-precision data support for the subsequent evaluation of the effectiveness of the monitoring plan, thereby helping decision makers to more accurately judge the actual effects of different monitoring plans in construction site management and make more reasonable decisions and adjustments.

[0110] In a possible implementation manner, determining the monitoring noise error corresponding to the construction behavior event of the monitoring execution data of each construction site node for the node scene feature of each construction site node includes:

[0111] Step A110, estimating the noise error weights corresponding to the construction behavior events of the monitoring execution data of the X construction site nodes respectively through a noise error estimation network based on the node scene features corresponding to the X construction site nodes respectively.

[0112] Step A120, calculating the mean of the noise error weights corresponding to the X construction site nodes respectively, to generate a reference noise error weight.

[0113] Step A130, obtaining the noise error weight corresponding to the a-th construction site node and the loss value between the reference noise error weight, and outputting the loss value as the monitoring noise error corresponding to the construction behavior event of the monitoring execution data of each construction site node for the node scene feature of the a-th construction site node. a is a positive integer not greater than X.

[0114] In this embodiment, the previous five construction site nodes (node ​​A, node B, node C, node D and node E) are still taken as an example.

[0115] The server first processes the node scene features corresponding to the five nodes through the noise error estimation network. It is assumed that the node scene features include the weather conditions at the construction site, the aging degree of the construction equipment, the experience level of the construction personnel, etc.

[0116] For node A, there was strong wind at the construction site, the construction equipment was old, and the construction workers were experienced. Based on these characteristics, the noise error estimation network estimated that the noise error weight corresponding to the construction behavior event of node A for the monitoring execution data was 0.4.

[0117] For node B, the ground at the construction site is uneven, the construction equipment is relatively new, and the construction workers are generally inexperienced. The network estimates that the noise error weight of node B is 0.3.

[0118] For node C, the humidity at the construction site is high, the construction equipment is under normal maintenance, and the construction personnel are inexperienced. The network estimates that the noise error weight of node C is 0.5.

[0119] For node D, the construction site temperature is high, the construction equipment occasionally fails, and the construction personnel are experienced. The network estimates the noise error weight of node D to be 0.45.

[0120] For node E, the lighting conditions at the construction site are poor, the construction equipment has good performance, and the construction personnel are experienced. The network estimates that the noise error weight of node E is 0.35.

[0121] The server then calculates the average of the noise error weights corresponding to the five construction site nodes, i.e. (0.4+0.3+0.5+0.45+0.35) / 5=0.4. The generated 0.4 is the baseline noise error weight.

[0122] Then, the loss value between the noise error weight corresponding to the a-th construction site node and the benchmark noise error weight is obtained.

[0123] Taking node A as an example, the noise error weight of node A is 0.4, and the baseline noise error weight is 0.4. The two are the same, and the loss value is 0.

[0124] For node B, its noise error weight is 0.3, compared with the baseline noise error weight of 0.4, the loss value is calculated as 0.4-0.3=0.1.

[0125] For node C, the noise error weight is 0.5 and the loss value is 0.5-0.4=0.1.

[0126] For node D, the noise error weight is 0.45 and the loss value is 0.45-0.4=0.05.

[0127] For node E, the noise error weight is 0.35 and the loss value is 0.4-0.35=0.05.

[0128] The server finally outputs these loss values ​​as the node scene features of the corresponding nodes for each construction site node and the monitoring noise error corresponding to the construction behavior events of the monitoring execution data.

[0129] In the actual processing process, the server needs to efficiently and accurately call the noise error estimation network to process a large amount of node scene feature data. At the same time, the server needs to perform precise weight mean calculation and loss value calculation to ensure the accuracy and reliability of the results.

[0130] When calculating the loss value, the server needs to follow the predetermined loss function and algorithm. For example, if the mean square error is used as the loss function, the server will strictly follow the formula for calculation.

[0131] The server will also monitor and verify the data in the calculation process in real time. If it is found that the noise error weight of a node is abnormal, or the loss value calculation result does not meet expectations, the server will immediately review and correct the error.

[0132] For large amounts of data processing, the server will reasonably allocate computing resources to ensure the efficiency and stability of the entire computing process. At the same time, the server will properly store the calculated noise error results for subsequent optimization of construction behavior image data.

[0133] For example, when subsequently optimizing the construction behavior image data of node B, the server will refer to the previously calculated loss value of 0.1 to determine the corresponding correction coefficient and optimization strategy.

[0134] Through such precise calculation and processing, the server can provide accurate monitoring noise error assessment for each construction site node, providing an important basis for further optimizing construction behavior image data and evaluating the effectiveness of monitoring solutions.

[0135] In a possible implementation, step S120 includes:

[0136] Step S121, obtaining the first construction behavior record data of the ath construction site node among the X construction site nodes for the configuration monitoring execution data corresponding to the ath construction site node, where a is a positive integer not greater than X.

[0137] Step S122, obtaining Y types of construction behavior characteristic parameters of the a-th construction site node corresponding to the configuration monitoring execution data from the first construction behavior record data corresponding to the a-th construction site node, where Y is a positive integer.

[0138] Step S123: extracting the real construction behavior image data of the a-th construction site node for the configuration monitoring execution data based on the Y types of construction behavior characteristic parameters of the a-th construction site node.

[0139] In this embodiment, the five construction site nodes (node ​​A, node B, node C, node D, and node E) mentioned above are still taken as an example, and X=5.

[0140] First, the server obtains the first construction behavior record data of each of the five construction site nodes for its corresponding configuration monitoring execution data.

[0141] For node A, the server obtains the first construction behavior record data corresponding to it. These data include various detailed records of node A performing construction tasks according to the test monitoring plan, such as the construction start time, the specific operation process of the construction personnel, and the relevant information of the materials and equipment used.

[0142] Next, the server obtains Y construction behavior characteristic parameters of node A for its corresponding configuration monitoring execution data from the first construction behavior record data corresponding to node A. Assume that Y=3, and these three construction behavior characteristic parameters are construction progress, construction quality, and construction material consumption.

[0143] The server screens and extracts the first construction behavior record data. For example, by comparing the completion time and planned time of each process in the construction process, the construction progress characteristic parameter value of node A is calculated to be 70%; through the results of quality inspection and evaluation of the completed construction part, the construction quality characteristic parameter value is obtained to be 90 points (out of 100 points); by statistics and calculations of the materials used, the construction material consumption characteristic parameter value is obtained to be 50 units.

[0144] Then, based on the three construction behavior characteristic parameters of node A, the server extracts the real construction behavior image data of node A for the configuration monitoring execution data.

[0145] The server integrates and processes these characteristic parameter values ​​to generate image data that can intuitively reflect the construction behavior of node A. For example, the construction progress of 70% is displayed in the form of a progress bar, the distribution of construction quality of 90 points is displayed in the form of a chart, and the consumption of construction materials of 50 units is displayed in the form of a bar chart.

[0146] For node B, the server also obtains its first construction behavior record data. Assume that the three extracted construction behavior characteristic parameters are the operation time of construction equipment, the work efficiency of construction personnel and the number of construction safety accidents.

[0147] In the first construction behavior record data, the server counts the startup and running time of the construction equipment and obtains the characteristic parameter value of the construction equipment running time as 8 hours; by calculating the workload completed by the construction personnel in unit time, the characteristic parameter value of the construction personnel work efficiency is 80%; by recording and counting the safety incidents during the construction process, the characteristic parameter value of the number of construction safety accidents is 0 times.

[0148] Based on these three construction behavior characteristic parameters, the server generates real construction behavior image data of node B, such as using a line graph to show the changing trend of construction equipment operation time, using a pie chart to show the composition ratio of construction personnel's work efficiency, and using numbers to directly display the number of construction safety accidents.

[0149] For node C, node D, and node E, the server also processes according to the same steps and methods.

[0150] Taking node C as an example, assuming that the three extracted construction behavior characteristic parameters are the number of completed construction processes, construction environment temperature, and construction noise level. The server accurately extracts and calculates the first construction behavior record data to obtain the corresponding characteristic parameter values, and then generates real construction behavior image data that can clearly present these characteristics.

[0151] For node D, it is assumed that the extracted feature parameters are the number of safety inspections, the number of safety hazards discovered, and the safety rectification completion rate.

[0152] For node E, it is assumed that the extracted feature parameters are the number of construction workers on duty, the number of construction workers on leave, and the length of overtime hours of construction workers.

[0153] Throughout the process, the server needs to quickly and accurately process large amounts of data and convert it into meaningful image data. The server must ensure that data extraction and calculation are accurate and that image data generation is clear and intuitive so that subsequent analysis and processing can be effective.

[0154] At the same time, the server must strictly monitor and manage the source and processing of data to ensure the security and integrity of the data. If an abnormality occurs during data processing, the server must be able to detect it in time and take appropriate measures to correct it.

[0155] Through detailed processing of each construction site node, the server provides rich and accurate real construction behavior image data for subsequent monitoring and analysis, which helps to better evaluate the construction situation and the execution effect of the monitoring plan.

[0156] In a possible implementation, step S122 includes:

[0157] Step S1221: Obtain, from the first construction behavior record data corresponding to the a-th construction site node, the construction behavior occurrence parameters of the a-th construction site node corresponding to its corresponding configuration monitoring execution data.

[0158] Step S1222: Obtain, from the first construction behavior record data corresponding to the a-th construction site node, a construction behavior duration parameter of the a-th construction site node corresponding to the configuration monitoring execution data.

[0159] Step S1223, outputting the construction behavior occurrence parameters and construction behavior duration parameters corresponding to the a-th construction site node as Y types of construction behavior characteristic parameters of the a-th construction site node for its corresponding configuration monitoring execution data.

[0160] Still taking the five construction site nodes mentioned earlier (node ​​A, node B, node C, node D and node E) as an example, suppose we focus on node A.

[0161] First, the server obtains the construction behavior occurrence parameters from the first construction behavior record data corresponding to node A.

[0162] The first construction behavior record data of node A records the information of various construction behaviors in detail. For example, regarding the construction behavior of "pouring concrete", the server obtains the specific time of its occurrence as 9 am, which is the occurrence parameter of the construction behavior.

[0163] For the construction activity of "installing scaffolding", the server obtained the time of occurrence as 2 pm.

[0164] Next, the server obtains the construction behavior duration parameter from the first construction behavior record data corresponding to node A.

[0165] Still taking "pouring concrete" as an example, the server calculates the duration of the construction behavior as 2 hours through the start time of 9 am and the end time of 11 am in the records. This is the duration parameter of the construction behavior "pouring concrete".

[0166] For the construction activity of "installing scaffolding", it started at 2 pm and ended at 4 pm, lasting for 2 hours.

[0167] Then, the server outputs the construction behavior occurrence parameters (9 a.m.) and construction behavior duration parameters (2 hours) of "pouring concrete" corresponding to node A, as well as the construction behavior occurrence parameters (2 p.m.) and construction behavior duration parameters (2 hours) of "installing scaffolding", etc., as Y types of construction behavior characteristic parameters of node A for its corresponding configuration monitoring execution data.

[0168] Assume that Y is 5. In addition to the two construction activities mentioned above, there are also "laying pipes", "wall plastering" and "electrical wiring".

[0169] For “Laying Pipelines”, the server obtains that the occurrence time is 10:00 AM and the duration is 1.5 hours.

[0170] For "Wall Plastering", the occurrence time is 3 pm and the duration is 1 hour.

[0171] For "Electrical Wiring", the occurrence time is 11 am and the duration is 2.5 hours.

[0172] The server organizes and outputs the occurrence parameters and duration parameters of these construction behaviors as Y types of construction behavior characteristic parameters of node A.

[0173] During the processing, the server needs to accurately parse a large number of complex data records. When extracting time information, it is necessary to ensure accuracy and avoid deviations caused by inconsistent data formats or recording errors.

[0174] At the same time, the server must classify and organize different types of construction behaviors to ensure that the characteristic parameters of each construction behavior can be correctly acquired and output.

[0175] In actual operation, if there is ambiguous or incomplete information in the first construction behavior record data, the server will try to supplement and verify it through other relevant data. For example, if the end time record of "wall plastering" is missing, the server may infer its end time based on the construction worker's punch-in record or the start time of the adjacent construction behavior, thereby calculating the accurate duration parameter.

[0176] In addition, the server will also conduct a preliminary rationality check on the acquired characteristic parameters. For example, if the duration of a construction behavior is obviously too long or too short, which is inconsistent with normal construction conditions, the server will issue an alarm and conduct further verification.

[0177] For multiple construction site nodes, the server will process them according to the same process and standards. For example, for node B, the server will also obtain similar construction behavior occurrence parameters and duration parameters from its first construction behavior record data, and output them as node B's construction behavior characteristic parameters.

[0178] Through such rigorous and meticulous processing, the server can provide accurate and reliable basic data for subsequent data analysis and monitoring program evaluation.

[0179] In a possible implementation, step S120 further includes:

[0180] Step S124, determining the linkage construction site node corresponding to the ath construction site node among the X construction site nodes. The linkage construction site node is a construction site node among the X construction site nodes that receives the derived monitoring execution data derived and extended by the ath construction site node. a is a positive integer not greater than X.

[0181] Step S125, obtaining the second construction behavior record data of the derived monitoring execution data derived and expanded by the linked construction site node for the a-th construction site node.

[0182] Step S126, determining, from the second construction behavior record data, Y types of construction behavior characteristic parameters of the derived monitoring execution data derived and extended by the linkage construction site node for the a-th construction site node as Y types of derived collaborative construction characteristic parameters corresponding to the a-th construction site node. Y is a positive integer.

[0183] Step S127, based on the Y types of derived collaborative construction feature parameters corresponding to the a-th construction site node, extract the real construction collaborative behavior image data of the derived monitoring execution data derived and extended from the a-th construction site node.

[0184] In this embodiment, the previous five construction site nodes (node ​​A, node B, node C, node D and node E) are still taken as an example, and X=5.

[0185] First, determine the linkage construction site node corresponding to the ath construction site node. Assuming we focus on node A (a=1), the change in the construction progress of node A may affect the construction arrangements of nodes B and C, so nodes B and C are the linkage construction site nodes corresponding to node A.

[0186] Next, the server obtains the second construction behavior record data of the derived monitoring execution data derived and expanded from Node A by the linked construction site nodes (Node B and Node C).

[0187] For node B, its second construction behavior record data may show that due to the early progress of node A, node B increased the input of construction materials and adjusted the work allocation of construction personnel accordingly. For node C, the record data may show that due to the change in the progress of node A, node C started the maintenance work of some construction equipment in advance.

[0188] Then, the server determines Y construction behavior characteristic parameters of the derived monitoring execution data derived and extended by nodes B and C for node A from these second construction behavior record data. Assuming Y=3, these three construction behavior characteristic parameters are the adjustment range of construction resources, the number of construction plan changes, and the degree of improvement of construction coordination efficiency.

[0189] For node B, the server obtains from its second construction behavior record data that due to the influence of node A, the adjustment range of construction resources is increased by 20%, the number of construction plan changes is 1, and the degree of improvement of construction coordination efficiency is 15%. For node C, the adjustment range of construction resources is increased by 10%, the number of construction plan changes is 0, and the degree of improvement of construction coordination efficiency is 10%.

[0190] These parameters serve as three derived collaborative construction characteristic parameters corresponding to node A.

[0191] Based on the three derived collaborative construction feature parameters corresponding to node A, the server extracts the real construction collaborative behavior image data of the derived monitoring execution data derived and extended by node A.

[0192] The server integrates and processes these feature parameters to generate image data that can intuitively reflect the synergistic impact of node A on node B and node C.

[0193] For example, a line graph is used to show the changes in the adjustment range of construction resources of nodes B and C; a bar graph is used to compare the number of construction plan changes of nodes B and C; and a radar chart is used to show the degree of improvement in the construction coordination efficiency of nodes B and C.

[0194] The same process applies to other nodes.

[0195] Assuming that we focus on node D (a=4), the potential safety hazard discovered by node D may cause node E to adjust the work arrangement of personnel. Node E is the linkage construction site node corresponding to node D.

[0196] The server obtains the second construction behavior record data of the derived monitoring execution data derived and expanded by node E for node D, and determines the characteristic parameters of the construction behavior therefrom, such as the number of personnel adjustments, changes in working hours, and the degree of strengthening of safety measures.

[0197] Based on these feature parameters, the server extracts the real construction collaborative behavior image data of the derived monitoring execution data derived and extended by node D.

[0198] Throughout the process, the server needs to accurately identify the linkage relationship between each node and accurately obtain the required information from a large amount of construction behavior record data. At the same time, the server needs to effectively process and analyze this data to generate clear and valuable construction collaborative behavior image data.

[0199] The server must also ensure the security and integrity of the data to avoid data loss or leakage. For any abnormal situation that occurs during data processing, the server must be able to detect it in time and take corresponding error correction measures.

[0200] By extracting the derived collaborative construction feature parameters of each node and generating image data, the server provides strong support for comprehensively evaluating the collaborative work effect and effectiveness of the monitoring scheme at the construction site.

[0201] In a possible implementation, the real construction behavior image data corresponding to each construction site node includes Y types of construction behavior characteristic parameters of each construction site node for configuration monitoring execution data, the optimized construction behavior image data corresponding to each construction site node includes Y types of optimized construction behavior characteristic parameters generated by optimizing the Y types of construction behavior characteristic parameters, the real construction collaborative behavior image data corresponding to each construction site node includes Y types of derived collaborative construction characteristic parameters of derived monitoring execution data of each construction site node, and the optimized construction collaborative behavior image data corresponding to each construction site node includes Y types of optimized derived collaborative construction characteristic parameters generated by optimizing the Y types of derived collaborative construction characteristic parameters.

[0202] The method further comprises:

[0203] Step B110, fuse the same optimized construction behavior characteristic parameter among the Y kinds of optimized construction behavior characteristic parameters corresponding to the construction site nodes in the test cluster, and generate the construction behavior characteristic global parameters corresponding to the Y kinds of optimized construction behavior characteristic parameters of the test cluster, and fuse the same optimized derived collaborative construction characteristic parameter among the Y kinds of optimized derived collaborative construction characteristic parameters corresponding to the construction site nodes in the test cluster, and generate the optimized derived collaborative construction characteristic global parameters corresponding to the Y kinds of optimized derived collaborative construction characteristic parameters of the test cluster.

[0204] Step B120, obtaining the construction behavior feature global parameters corresponding to the Y optimized construction behavior feature parameters of the reference cluster, and the optimized derived collaborative construction feature global parameters corresponding to the Y optimized derived collaborative construction feature parameters of the reference cluster.

[0205] Step B130, performing loss calculation on the global parameters of the construction behavior characteristics corresponding to the same optimized construction behavior characteristic parameters in the test cluster and the reference cluster, generating first loss calculation results corresponding to the Y types of optimized construction behavior characteristic parameters respectively, and outputting the Y types of first loss calculation results as the construction behavior state error information.

[0206] Step B140, performs loss calculation on the optimized derived collaborative construction feature global parameters corresponding to the same optimized derived collaborative construction feature parameters in the test cluster and the reference cluster, generates second loss calculation results corresponding to the Y types of optimized derived collaborative construction feature parameters respectively, and outputs the Y types of second loss calculation results as the collaborative behavior state error information.

[0207] In this embodiment, the five construction site nodes mentioned above (node ​​A, node B, node C belong to the test cluster, and node D and node E belong to the reference cluster) are still taken as an example.

[0208] Assuming Y=3, the three construction behavior characteristic parameters are construction progress, construction quality and construction resource consumption; the three derived collaborative construction characteristic parameters are collaborative efficiency, collaborative resource adjustment and collaborative quality impact.

[0209] For node A, its real construction behavior image data includes three construction behavior characteristic parameters for configuration monitoring execution data, such as construction progress of 70%, construction quality of 90 points (out of 100 points), and construction resource consumption of 50 units. After optimization, the optimized construction behavior image data of node A includes three optimized construction behavior characteristic parameters, such as optimized construction progress of 75%, optimized construction quality of 95 points, and optimized construction resource consumption of 45 units.

[0210] The real construction collaborative behavior image data of node A includes three derived collaborative construction characteristic parameters of its derived monitoring execution data, such as collaborative efficiency of 80%, collaborative resource adjustment of 10 units, and collaborative quality impact of 5 points. After optimization, the optimized construction collaborative behavior image data of node A includes three optimized derived collaborative construction characteristic parameters, such as the optimized collaborative efficiency of 85%, the optimized collaborative resource adjustment of 15 units, and the optimized collaborative quality impact of 8 points.

[0211] For node B and node C, there are also similar real construction behavior characteristic parameters, optimized construction behavior characteristic parameters, real construction collaborative behavior characteristic parameters and optimized construction collaborative behavior characteristic parameters.

[0212] The server performs fusion processing on the same optimized construction behavior characteristic parameter among the three optimized construction behavior characteristic parameters corresponding to the construction site nodes (node ​​A, node B, node C) in the test cluster.

[0213] Taking the construction progress as an example, assuming that the optimized construction progress of node B is 80% and the optimized construction progress of node C is 75%, the server integrates the optimized construction progress of these three nodes, such as calculating the average value, and obtains the global parameter of the construction behavior characteristic corresponding to the construction progress of the test cluster as (75%+80%+75%) / 3=76.67%.

[0214] Similarly, the two optimized construction behavior characteristic parameters, construction quality and construction resource consumption, are fused to generate the corresponding construction behavior characteristic global parameters of the test cluster.

[0215] For the optimized derived collaborative construction feature parameters, taking collaborative efficiency as an example, assuming that the optimized collaborative efficiency of node B is 90% and the optimized collaborative efficiency of node C is 85%, the server performs fusion processing to obtain the optimized derived collaborative construction feature global parameters corresponding to the collaborative efficiency of the test cluster.

[0216] In the same way, the server obtains the construction behavior feature global parameters corresponding to the three optimized construction behavior feature parameters of the reference cluster (node ​​D, node E), as well as the optimized derived collaborative construction feature global parameters corresponding to the three optimized derived collaborative construction feature parameters.

[0217] Next, the server calculates the loss of the global parameters of the construction behavior characteristics corresponding to the same optimized construction behavior characteristic parameters in the test cluster and the reference cluster.

[0218] Taking the construction progress as an example, assuming that the global parameter of the construction behavior characteristic corresponding to the construction progress of the reference cluster is 70%, and that of the test cluster is 76.67%, the server calculates the losses of the two, such as using the mean square error calculation, to obtain the first loss calculation result corresponding to the construction progress, the optimized construction behavior characteristic parameter.

[0219] The same loss calculation is performed on the construction quality and the construction resource consumption to generate the first loss calculation results corresponding to the two optimized construction behavior characteristic parameters, and the three first loss calculation results are output as the construction behavior state error information.

[0220] For the optimized derived collaborative construction feature parameters, taking collaborative efficiency as an example, assuming that the optimized derived collaborative construction feature global parameter corresponding to the collaborative efficiency of the reference cluster is 80%, and that of the test cluster is 85%, the server performs loss calculation to obtain the second loss calculation result corresponding to the optimized derived collaborative construction feature parameter, collaborative efficiency.

[0221] The same calculation is performed on collaborative resource adjustment and collaborative quality impact to generate the second loss calculation results corresponding to these two optimization-derived collaborative construction characteristic parameters, and these three second loss calculation results are output as collaborative behavior state error information.

[0222] Throughout the entire processing process, the server needs to accurately perform a large amount of data calculation and fusion operations. For the various characteristic parameters of each construction site node, the server must accurately extract, optimize and classify them. During the fusion and loss calculation stages, the server must strictly follow the predetermined algorithms and rules to ensure the accuracy and reliability of the results.

[0223] The server also needs to monitor and verify the data in real time during the processing. If the calculation result of a feature parameter is found to be abnormal, or the global parameter after fusion processing does not meet expectations, the server will conduct a backtracking check to find out the problem and correct it.

[0224] Through comprehensive processing and comparative analysis of the characteristic parameters of the test cluster and the reference cluster, the server can provide comprehensive and accurate error information for evaluating the effectiveness of the monitoring scheme, thereby providing strong support for decision-making.

[0225] In a possible implementation, there is a correlation between one of the Y first loss calculation results and one of the Y second loss calculation results.

[0226] Step S140 includes:

[0227] Step S141, based on the Y types of first loss calculation results and the Y types of second loss calculation results, weighted calculation is performed on the associated first loss calculation results and the second loss calculation results to generate Y first scheme measurement values ​​of the test monitoring scheme and the reference monitoring scheme for the construction behavior.

[0228] Step S142: Determine the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme respectively based on the Y first scheme metric values.

[0229] In this embodiment, the previous five construction site nodes are still taken as an example, assuming that Y=3, that is, there are three construction behavior characteristic parameters and three derived collaborative construction characteristic parameters, corresponding to three first loss calculation results and three second loss calculation results.

[0230] It is assumed that these three are the first loss calculation results of construction progress, construction quality and construction resource consumption, and the second loss calculation results of collaborative efficiency, collaborative resource adjustment and collaborative quality impact.

[0231] In actual situations, the calculation result of the first loss of construction progress may be correlated with the calculation result of the second loss of collaborative efficiency. For example, if the construction progress is fast, it may promote the improvement of collaborative efficiency; on the contrary, if the construction progress lags behind, it may affect the collaborative efficiency.

[0232] The server performs weighted calculation on the results with correlation based on the three first loss calculation results and the three second loss calculation results.

[0233] Assume that the first loss calculation result of construction progress is 0.2, the second loss calculation result of collaborative efficiency is 0.15, and their associated weights are 0.6 and 0.4 respectively. Then the weighted calculation result is 0.2*0.6+0.15*0.4=0.18, which is the first solution metric value generated for the association relationship between construction progress and collaborative efficiency.

[0234] Similarly, for the two pairs of loss calculation results that are associated with each other, namely, construction quality and collaborative resource adjustment, and construction resource consumption and collaborative quality impact, the server also performs weighted calculations according to their respective associated weights to generate two other first solution metrics.

[0235] Based on the Y (ie, 3) first solution metric values, the server determines the monitoring effectiveness corresponding to the test monitoring solution and the reference monitoring solution, respectively.

[0236] For example, if the three first solution metric values ​​are 0.18, 0.22, and 0.16 respectively.

[0237] The server first obtains the number of the first solution metric values ​​that are greater than a first threshold value (assuming it is 0.15). In this example, all three metric values ​​are greater than 0.15.

[0238] Then, the server determines whether the number greater than the first threshold is not less than the second threshold (assuming it is 2). In this example, the number greater than the first threshold is 3, which is greater than the second threshold of 2.

[0239] Since the number greater than the first threshold value is not less than the second threshold value, the server determines that the monitoring effectiveness corresponding to the test monitoring scheme is better than the monitoring effectiveness corresponding to the reference monitoring scheme.

[0240] During the whole process, the server needs to accurately identify and process the correlation between various loss calculation results and perform weighted calculations according to reasonable weights. At the same time, the server must objectively and accurately judge the effectiveness of the monitoring plan based on the set threshold value and quantity standard.

[0241] The server also strictly manages and monitors the data during the calculation and judgment process to ensure the accuracy and reliability of the data. If an abnormal situation occurs during the calculation or judgment process, the server will promptly investigate and handle it to ensure the fairness and effectiveness of the final result.

[0242] Through such detailed and comprehensive analysis and judgment, the server can provide a scientific and reliable basis for the selection of monitoring solutions for construction sites, which helps to improve the efficiency and quality of construction management.

[0243] In a possible implementation, step S142 includes:

[0244] Step S1421: Obtain the number of first solution metric values ​​greater than a first threshold value among the Y first solution metric values.

[0245] Step S1422: If the number is not less than the second threshold value, it is determined that the monitoring effectiveness corresponding to the test monitoring scheme is better than the monitoring effectiveness corresponding to the reference monitoring scheme.

[0246] Step S1423: If the number is less than the second threshold value, it is determined that the monitoring effectiveness corresponding to the reference monitoring scheme is better than the monitoring effectiveness corresponding to the test monitoring scheme.

[0247] In this embodiment, still taking the previous example, it is assumed that Y=3, that is, there are three first solution metric values, namely solution metric value 1, solution metric value 2 and solution metric value 3.

[0248] Assume that the first threshold value is set to 0.8 and the second threshold value is set to 2.

[0249] The server first obtains the three first solution metric values ​​and compares them with the first threshold value 0.8 respectively.

[0250] Assume that the metric value of solution 1 is 0.9, the metric value of solution 2 is 0.7, and the metric value of solution 3 is 0.85.

[0251] By comparison, the server determines that the first solution metric values ​​greater than the first threshold value 0.8 include solution metric value 1 and solution metric value 3, which are 2 in total.

[0252] Then, the server determines the number of first solution metric values ​​greater than the first threshold value, which is 2 in this example.

[0253] Next, the server compares this number with a second threshold value 2.

[0254] Since 2 is equal to the second threshold value 2, the condition that the quantity is not less than the second threshold value is satisfied.

[0255] Based on this, the server determines that the monitoring effectiveness corresponding to the test monitoring scheme is better than the monitoring effectiveness corresponding to the reference monitoring scheme.

[0256] Assume another situation, the three first solution metric values ​​are solution metric value 1 is 0.85, solution metric value 2 is 0.75, and solution metric value 3 is 0.7.

[0257] After comparison, the server determines that the only first solution metric value greater than the first threshold value 0.8 is solution metric value 1, and the number is 1.

[0258] Because 1 is less than the second threshold value 2, the server determines that the monitoring effectiveness corresponding to the reference monitoring solution is better than the monitoring effectiveness corresponding to the test monitoring solution.

[0259] During the entire processing, the server needs to accurately compare each first solution metric with the first threshold value and accurately count the number of metrics greater than the first threshold value, which requires the server to have efficient computing power and accurate data processing capabilities.

[0260] The server must also strictly follow the set second threshold value to judge the effectiveness of the monitoring scheme. In the comparison and judgment process, the server cannot make any mistakes, otherwise it will lead to misjudgment of the monitoring effectiveness, which will affect the subsequent construction management decision.

[0261] At the same time, the server will record and store the results of each comparison and judgment for subsequent query and analysis. If data anomalies or calculation errors occur during the processing, the server can trace back these records to find the problem and make corrections.

[0262] Through such a rigorous and accurate processing flow, the server can provide construction management with reliable judgments on the effectiveness of monitoring plans, help managers choose more suitable monitoring plans, and improve the management level and construction quality of the construction site.

[0263] In a possible implementation, the method further includes:

[0264] Step C110, adding Z construction site nodes to the verification cluster and the comparison cluster, configuring third monitoring execution data to the construction site nodes in the verification cluster according to the test monitoring scheme, and configuring fourth monitoring execution data to the construction site nodes in the comparison cluster according to the reference monitoring scheme. Z is a positive integer greater than 1.

[0265] Step C120, based on the optimized construction behavior image data of the construction site nodes in the verification cluster, the optimized construction behavior image data of the construction site nodes in the comparison cluster, the optimized construction collaborative behavior image data of the construction site nodes in the verification cluster, and the optimized construction collaborative behavior image data of the construction site nodes in the comparison cluster, determine the Y second scheme measurement values ​​for the operation of the test monitoring scheme and the reference monitoring scheme.

[0266] Step S142 includes: determining the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme respectively based on the Y first scheme metric values ​​and the Y second scheme metric values.

[0267] Assume that Z=6, the newly added 6 construction site nodes are node F, node G, node H, node I, node J and node K.

[0268] The server adds the six construction site nodes to the verification cluster and the comparison cluster respectively. Assume that nodes F, G, and H are assigned to the verification cluster, and nodes I, J, and K are assigned to the comparison cluster.

[0269] According to the test monitoring plan, the server configures the third monitoring execution data to the construction site nodes (nodes F, G, and H) in the verification cluster. For example, node F is required to submit an inventory report of construction materials every 3 hours and record the loss of materials in detail; node G is required to monitor the wearing of safety protection equipment by construction workers in real time; and node H is required to summarize the optimization points and problems encountered in the construction process every day.

[0270] According to the reference monitoring scheme, the server configures the fourth monitoring execution data to the construction site nodes (node ​​I, node J and node K) in the comparison cluster. For example, node I is required to conduct a routine inspection of the construction equipment every 5 hours and record the operating status of the equipment; node J is required to evaluate the work efficiency and team collaboration of the construction team every week; node K is required to report the impact of environmental changes on the construction site on the construction in a timely manner.

[0271] In subsequent processing, the server obtains the optimized construction behavior image data and optimized construction collaborative behavior image data of the construction site nodes (node ​​F, node G, and node H) within the verification cluster, and compares the optimized construction behavior image data and optimized construction collaborative behavior image data of the construction site nodes (node ​​I, node J, and node K) within the cluster.

[0272] Assume that through a series of calculations and processing, the server determines the second solution metric values ​​for the three aspects of construction progress, construction quality, and construction safety. Taking the construction progress as an example, the server calculates that the average construction progress in the verification cluster is 80%, and the average construction progress in the comparison cluster is 70%, and then obtains the second solution metric value for the construction progress. Similarly, for construction quality and construction safety, the corresponding second solution metric values ​​are also calculated.

[0273] When determining the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme respectively based on the Y first scheme metric values ​​and the Y second scheme metric values, the server comprehensively considers these metric values.

[0274] Assume that Y=3, that is, there are 3 first solution measurement values ​​and 3 second solution measurement values.

[0275] The server first summarizes and analyzes these metrics. For each aspect of the metric (such as construction progress, construction quality, construction safety, etc.), the server combines the first solution metric with the second solution metric for evaluation.

[0276] For example, in terms of construction progress, suppose the first scheme metric shows that the test monitoring scheme is slightly better than the reference monitoring scheme, and the second scheme metric also shows that the test monitoring scheme is more effective in controlling the construction progress. The server will combine these two results to further strengthen the judgment on the effectiveness of the test monitoring scheme in terms of construction progress.

[0277] For construction quality, perhaps the first solution metric shows that there is little difference between the two, but the second solution metric shows that the reference monitoring solution has certain advantages. The server will comprehensively weigh these two different information and give a relatively objective judgment on the effectiveness of construction quality monitoring.

[0278] After combining all the metrics, the server will form a comprehensive and systematic evaluation result.

[0279] If the metric values ​​corresponding to the test monitoring solution are better than those of the reference monitoring solution in most aspects (for example, more than half), the server will ultimately determine that the monitoring effectiveness of the test monitoring solution is better. Conversely, if the reference monitoring solution performs better in most aspects, the server will determine that the monitoring effectiveness of the reference monitoring solution is better.

[0280] Throughout the entire process, the server needs to accurately process and analyze a large amount of data, including monitoring execution data of each construction site node, construction behavior image data, and collaborative behavior image data, etc. At the same time, the server must use complex algorithms and evaluation models to ensure that the judgment on the effectiveness of the monitoring plan is accurate and reliable.

[0281] The server also needs to strictly monitor and manage the data processing and evaluation process to promptly detect and correct possible errors. If new information or data changes appear during the evaluation process, the server can flexibly adjust the evaluation strategy to ensure the fairness and effectiveness of the final result.

[0282] Through such a comprehensive and in-depth comprehensive evaluation, the server can provide the most suitable monitoring solution for the construction site, thereby improving the level and quality of construction management and ensuring the smooth progress of the construction project.

[0283] In a possible implementation, the method further includes:

[0284] Step D110, based on the connection topology graph between the Q construction site nodes, the Q construction site nodes are divided to generate W node groups. The connection topology graph represents the connection communication relationship between the Q construction site nodes, and the construction site nodes in each node group in the W node groups have communication connections, wherein Q is an integer greater than 1, and W is a positive integer not greater than Q.

[0285] Step D120: extract R node groups that meet the test criteria from the W node groups, and output the construction site nodes in the R node groups as the X construction site nodes. R is a positive integer.

[0286] In this embodiment, it is assumed that Q=20, that is, there are 20 construction site nodes, which are marked as node 1 to node 20 respectively.

[0287] The server obtains the connection topology diagram between the 20 construction site nodes. This connection topology diagram clearly shows the connection and communication relationship between each node. For example, node 1 has a direct communication connection with node 2 and node 3; node 4 is connected with node 5 and node 6, and so on.

[0288] Based on this connection topology, the server divides the nodes into W node groups by analyzing factors such as the connection density between nodes, communication frequency, and the relevance of construction tasks.

[0289] Assume that after division, five node groups are generated, namely:

[0290] Group 1: includes Node 1, Node 2, Node 3, Node 4, and Node 5. These nodes are in the same construction area, communicate frequently with each other, and are jointly responsible for the construction of a basic project.

[0291] Group 2: includes nodes 6, 7, and 8. They are mainly responsible for the construction of the building structure and need to closely collaborate and communicate with each other.

[0292] Group 3: includes nodes 9, 10, 11, and 12. These nodes are involved in the installation of the electrical system and have close communication links.

[0293] Group 4: includes nodes 13, 14, and 15. They focus on pipeline laying and have a specific communication mode.

[0294] Group 5: includes nodes 16, 17, 18, 19, and 20. These nodes are engaged in decoration work and require constant coordination and communication.

[0295] Next, the server needs to extract R node groups that meet the test criteria from the five node groups. It is assumed that the test criteria include the complexity of the construction task, the key stages of the construction progress, and the degree of investment of construction resources.

[0296] Taking the complexity of construction tasks as an example, the basic engineering construction tasks of server evaluation group 1 are relatively complex and meet the test standards; the building structure construction tasks of group 2 are also complex and meet the standards. However, the electrical system installation tasks of group 3 are moderately complex and do not meet the higher requirements of the test standards.

[0297] For the critical stages of the construction progress, it is assumed that the pipe laying work of Group 4 is in a critical period, which is of great significance to the evaluation of the monitoring plan and meets the test standards; the decoration work of Group 5 has not yet entered the critical stage and does not meet the standards.

[0298] After comprehensive evaluation, the server determines that the three node groups, Group 1, Group 2, and Group 4, meet the test criteria, that is, R=3.

[0299] Then, the server outputs the construction site nodes in the three node groups that meet the criteria as X construction site nodes. In this example, group 1 has 5 nodes, group 2 has 3 nodes, and group 4 has 3 nodes, for a total of 11 nodes, that is, X=11.

[0300] During the entire processing process, the server needs to deeply analyze various information in the connection topology map and use complex algorithms and rules to divide the node groups. At the same time, for the evaluation of the test standards, the server needs to comprehensively consider multiple factors to ensure that the extracted node groups are representative and targeted.

[0301] The server also needs to repeatedly check and verify the division and extraction results to ensure that the output construction site nodes can be accurately and effectively used for subsequent monitoring plan evaluation and optimization work.

[0302] When processing large amounts of data and complex relationships, the server must maintain efficient and stable performance to avoid errors or delays. If the characteristics of certain node groups are found to be unclear or not in line with expectations during the process, the server will re-evaluate and adjust.

[0303] Through such a rigorous and meticulous processing flow, the server can select the most suitable nodes from numerous construction site nodes, provide strong support for the effectiveness evaluation of the monitoring plan, and thus promote the optimization and improvement of construction management.

[0304] Figure 2In an embodiment of the present application, a construction site information verification and processing system 100 based on image analysis is provided, including a processor 1001 and a memory 1003 and a sensor node code stored in the memory 1003. The processor 1001 executes the above-mentioned sensor node code to implement the steps of the sensor data processing method applied to smart photovoltaics.

[0305] Figure 2 The construction site information verification and processing system 100 based on image analysis shown includes: a processor 1001 and a memory 1003. Among them, the processor 1001 and the memory 1003 are connected, such as through a bus 1002. Optionally, the construction site information verification and processing system 100 based on image analysis may also include a transceiver 1004, and the transceiver 1004 may be used for data interaction between the server and other servers, such as data transmission and / or data reception. It should be noted that in actual scheduling, the transceiver 1004 is not limited to one, and the structure of the construction site information verification and processing system 100 based on image analysis does not constitute a limitation on the embodiments of the present application.

[0306] Processor 1001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 1001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0307] The bus 1002 may include a path to transmit information between the above components. The bus 1002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 1002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 2 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0308] The memory 1003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), magnetic disk storage medium, other magnetic storage devices, or any other medium that can be used to carry or store sensor node codes and can be read by a computer, without limitation herein.

[0309] The memory 1003 is used to store the sensor node code for executing the embodiment of the present application, and the execution is controlled by the processor 1001. The processor 1001 is used to execute the sensor node code stored in the memory 1003 to implement the steps shown in the above method embodiment.

[0310] The server includes but is not limited to: mobile terminals such as mobile phones, notebook computers, PADs, etc. and fixed terminals such as digital TVs, desktop computers, etc.

[0311] An embodiment of the present application provides a computer-readable storage medium, on which a sensor node code is stored. When the sensor node code is executed by a processor, the steps and corresponding contents of the aforementioned method embodiment can be implemented.

[0312] It should be understood that, although each operation step is indicated by arrows in the flowchart of the embodiment of the present application, the implementation order of these steps is not limited to the order indicated by the arrows. Unless there is clear explanation in this article, in some implementation scenarios of the embodiment of the present application, the implementation steps in each flowchart can be performed in other orders according to demand. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages according to actual implementation scenarios, and some or all of these sub-steps or stages may be executed at the same time, and each sub-step or stage in these sub-steps or stages may also be executed at different times respectively. In different scenarios at the execution time, the execution order of these sub-steps or stages can be flexibly configured based on demand, and the embodiment of the present application does not limit this.

[0313] The above is only an optional implementation method for some implementation scenarios of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical concept of the scheme of the present application, other similar implementation methods based on the technical ideas of the present application are also within the protection scope of the embodiments of the present application.

Claims

1. A construction site information verification and processing method based on image analysis, characterized in that: The method includes: Add X construction site nodes to a test cluster and a reference cluster, configure first monitoring execution data to the construction site nodes in the test cluster according to a test monitoring scheme, and configure second monitoring execution data to the construction site nodes in the reference cluster according to a reference monitoring scheme; X is an integer greater than 1; Acquire real construction behavior image data of each construction site node among the X construction site nodes for the configured monitoring execution data, and acquire real construction collaborative behavior image data of the derived monitoring execution data derived and extended from each construction site node; the configured monitoring execution data is the monitoring execution data sent to each construction site node after configuration, and both the configured monitoring execution data and the derived monitoring execution data belong to the time series formed by the first monitoring execution data and the second monitoring execution data; Determine the node scene feature of each construction site node for the monitoring noise error corresponding to the construction behavior event of the monitoring execution data of each construction site node, optimize the real construction behavior image data and the real construction collaborative behavior image data corresponding to the X construction site nodes respectively based on the monitoring noise errors corresponding to the X construction site nodes respectively, and generate the optimized construction behavior image data and the optimized construction collaborative behavior image data corresponding to the X construction site nodes respectively; Based on the optimized construction behavior image data of the construction site nodes in the test cluster and the construction behavior state error information between the optimized construction behavior image data of the construction site nodes in the reference cluster, as well as the optimized construction collaborative behavior image data of the construction site nodes in the test cluster and the collaborative behavior state error information between the optimized construction collaborative behavior image data of the construction site nodes in the reference cluster, the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme are determined respectively.

2. The construction site information verification processing method based on image analysis according to claim 1 is characterized in that: The method optimizes the real construction behavior image data and the real construction collaborative behavior image data corresponding to the X construction site nodes based on the monitoring noise errors corresponding to the X construction site nodes, and generates the optimized construction behavior image data and the optimized construction collaborative behavior image data corresponding to the X construction site nodes, including: Based on the monitoring noise error corresponding to the a-th construction site node among the X construction site nodes, determine the noise correction coefficient corresponding to the a-th construction site node; a is a positive integer not greater than X; Performing mean calculation on the real construction behavior image data of the remaining construction site nodes to generate benchmark construction behavior image data, and performing mean calculation on the real construction collaborative behavior image data of the remaining construction site nodes to generate benchmark construction collaborative behavior image data; the remaining construction site nodes are the construction site nodes other than the a-th construction site node in the cluster where the a-th construction site node is located; Obtaining a first image feature difference between the real construction behavior image data corresponding to the a-th construction site node and the benchmark construction behavior image data, and a second image feature difference between the real construction collaborative behavior image data corresponding to the a-th construction site node and the benchmark construction collaborative behavior image data; According to the noise correction coefficient corresponding to the a-th construction site node, the first image feature difference is subjected to noise correction to generate a first corrected image feature difference; according to the noise correction coefficient corresponding to the a-th construction site node, the second image feature difference is subjected to noise correction to generate a second corrected image feature difference; The first corrected image feature difference is fused with the benchmark construction behavior image data to generate the optimized construction behavior image data corresponding to the a-th construction site node, and the second corrected image feature difference is fused with the benchmark construction collaborative behavior image data to generate the optimized construction collaborative behavior image data corresponding to the a-th construction site node.

3. The construction site information verification and processing method based on image analysis according to claim 1 is characterized in that: The determining of the monitoring noise error corresponding to the construction behavior event of the monitoring execution data of each construction site node by the node scene feature of each construction site node includes: By using a noise error estimation network, based on the node scene features respectively corresponding to the X construction site nodes, the noise error weights corresponding to the construction behavior events of the monitoring execution data of the X construction site nodes are estimated; Calculate the mean of the noise error weights corresponding to the X construction site nodes to generate a benchmark noise error weight; Obtain the noise error weight corresponding to the a-th construction site node, and the loss value between it and the baseline noise error weight, and output the loss value as the monitoring noise error corresponding to the construction behavior event of the monitoring execution data of each construction site node based on the node scene feature of the a-th construction site node; a is a positive integer not greater than X.

4. The construction site information verification processing method based on image analysis according to claim 1 is characterized in that: The obtaining of real construction behavior image data of each of the X construction site nodes for configuration monitoring execution data includes: Obtaining the first construction behavior record data of the a-th construction site node among the X construction site nodes for the configuration monitoring execution data corresponding to the a-th construction site node; a is a positive integer not greater than X; Obtaining, from the first construction behavior record data corresponding to the a-th construction site node, construction behavior occurrence parameters of the a-th construction site node corresponding to the configuration monitoring execution data; Obtaining, from the first construction behavior record data corresponding to the a-th construction site node, a construction behavior duration parameter of the a-th construction site node for the configuration monitoring execution data corresponding thereto; Output the construction behavior occurrence parameter and the construction behavior duration parameter corresponding to the a-th construction site node as Y types of construction behavior characteristic parameters of the a-th construction site node for the configuration monitoring execution data corresponding to the a-th construction site node; Y is a positive integer; Based on the Y types of construction behavior characteristic parameters of the a-th construction site node, the real construction behavior image data of the a-th construction site node for the configuration monitoring execution data is extracted.

5. The construction site information verification and processing method based on image analysis according to claim 1 is characterized in that: The step of obtaining the real construction collaborative behavior image data of the derived monitoring execution data derived and expanded from each construction site node includes: Determine a linkage construction site node corresponding to the a-th construction site node among the X construction site nodes; the linkage construction site node is a construction site node among the X construction site nodes that receives the derived monitoring execution data derived and extended by the a-th construction site node; a is a positive integer not greater than X; Obtaining second construction behavior record data of the derived monitoring execution data derived and extended by the linkage construction site node for the a-th construction site node; Determine, from the second construction behavior record data, Y types of construction behavior characteristic parameters of the derived monitoring execution data derived and extended by the linkage construction site node for the a-th construction site node as Y types of derived collaborative construction characteristic parameters corresponding to the a-th construction site node; Y is a positive integer; Based on the Y types of derived collaborative construction feature parameters corresponding to the a-th construction site node, the real construction collaborative behavior image data of the derived monitoring execution data derived and extended from the a-th construction site node is extracted.

6. The construction site information verification and processing method based on image analysis according to claim 1 is characterized in that: The real construction behavior image data corresponding to each construction site node includes Y types of construction behavior characteristic parameters of each construction site node for configuration monitoring execution data, the optimized construction behavior image data corresponding to each construction site node includes Y types of optimized construction behavior characteristic parameters generated by optimizing the Y types of construction behavior characteristic parameters, the real construction collaborative behavior image data corresponding to each construction site node includes Y types of derived collaborative construction characteristic parameters of the derived monitoring execution data of each construction site node, and the optimized construction collaborative behavior image data corresponding to each construction site node includes Y types of optimized derived collaborative construction characteristic parameters generated by optimizing the Y types of derived collaborative construction characteristic parameters; The method further comprises: The same optimized construction behavior characteristic parameter among the Y kinds of optimized construction behavior characteristic parameters corresponding to the construction site nodes in the test cluster is fused to generate the construction behavior characteristic global parameters corresponding to the Y kinds of optimized construction behavior characteristic parameters of the test cluster, and the same optimized derived collaborative construction characteristic parameter among the Y kinds of optimized derived collaborative construction characteristic parameters corresponding to the construction site nodes in the test cluster is fused to generate the optimized derived collaborative construction characteristic global parameters corresponding to the Y kinds of optimized derived collaborative construction characteristic parameters of the test cluster; Obtaining construction behavior feature global parameters corresponding to Y types of optimized construction behavior feature parameters of the reference cluster, and optimized derived collaborative construction feature global parameters corresponding to Y types of optimized derived collaborative construction feature parameters of the reference cluster; Performing loss calculation on the construction behavior feature global parameters corresponding to the same optimized construction behavior feature parameters in the test cluster and the reference cluster, generating first loss calculation results corresponding to the Y types of optimized construction behavior feature parameters respectively, and outputting the Y types of first loss calculation results as the construction behavior state error information; Loss calculation is performed on the optimized derived collaborative construction feature global parameters corresponding to the same optimized derived collaborative construction feature parameters in the test cluster and the reference cluster, and second loss calculation results corresponding to the Y types of optimized derived collaborative construction feature parameters are generated respectively, and the Y types of second loss calculation results are output as the collaborative behavior state error information.

7. The construction site information verification and processing method based on image analysis according to claim 6 is characterized in that: There is a correlation relationship between one of the Y first loss calculation results and one of the Y second loss calculation results; The method determines the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme respectively based on the construction behavior state error information between the optimized construction behavior image data of the construction site nodes in the test cluster and the optimized construction behavior image data of the construction site nodes in the reference cluster, and the collaborative behavior state error information between the optimized construction collaborative behavior image data of the construction site nodes in the test cluster and the optimized construction collaborative behavior image data of the construction site nodes in the reference cluster, including: Based on the Y first loss calculation results and the Y second loss calculation results, weighted calculation is performed on the associated first loss calculation results and the second loss calculation results to generate Y first scheme metric values ​​of the test monitoring scheme and the reference monitoring scheme for the construction behavior; Obtaining the number of first solution metric values ​​greater than a first threshold value among the Y first solution metric values; If the number is not less than the second threshold value, determining that the monitoring effectiveness corresponding to the test monitoring scheme is better than the monitoring effectiveness corresponding to the reference monitoring scheme; If the number is less than the second threshold value, it is determined that the monitoring effectiveness corresponding to the reference monitoring scheme is better than the monitoring effectiveness corresponding to the test monitoring scheme.

8. The construction site information verification and processing method based on image analysis according to claim 7 is characterized in that: The method further comprises: Add Z construction site nodes to the verification cluster and the comparison cluster, configure third monitoring execution data to the construction site nodes in the verification cluster according to the test monitoring scheme, and configure fourth monitoring execution data to the construction site nodes in the comparison cluster according to the reference monitoring scheme; Z is a positive integer greater than 1; Determine Y second scheme metric values ​​for the test monitoring scheme and the reference monitoring scheme for the operation based on the optimized construction behavior image data of the construction site nodes in the verification cluster, the optimized construction behavior image data of the construction site nodes in the comparison cluster, the optimized construction collaborative behavior image data of the construction site nodes in the verification cluster, and the optimized construction collaborative behavior image data of the construction site nodes in the comparison cluster; The determining, based on the Y first scheme metric values, the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme respectively includes: Based on the Y first scheme metric values ​​and the Y second scheme metric values, the monitoring effectiveness corresponding to the test monitoring scheme and the reference monitoring scheme respectively is determined.

9. The construction site information verification and processing method based on image analysis according to claim 1, characterized in that: The method further comprises: Based on a connection topology diagram between Q construction site nodes, the Q construction site nodes are divided to generate W node groups; the connection topology diagram represents the connection and communication relationship between the Q construction site nodes, and the construction site nodes in each node group in the W node groups have a communication connection, wherein Q is an integer greater than 1, and W is a positive integer not greater than Q; From the W node groups, R node groups that meet the test criteria are extracted, and the construction site nodes in the R node groups are output as the X construction site nodes; R is a positive integer.

10. A construction site information verification and processing system based on image analysis, characterized in that: The construction site information verification and processing system based on image analysis includes a processor and a machine-readable storage medium, in which machine-executable instructions are stored. The machine-executable instructions are loaded and executed by the processor to implement the construction site information verification and processing method based on image analysis of any one of claims 1-9.

Images