Bridge intelligent construction monitoring method and system

Abstract

The application is suitable for the technical field of construction monitoring, and particularly relates to a bridge intelligent construction monitoring method and system.The method comprises the following steps: drawing a panoramic image of a bridge construction site and dividing it into several subareas, setting the number of each subarea based on a preset numbering rule, collecting the construction tasks of construction personnel, defining the corresponding subarea as a task area, defining the monitoring period of the bridge construction, issuing a work reporting window to the construction personnel when the monitoring period arrives, and embedding a work reporting rule; reading the work reporting video uploaded by the construction personnel, and identifying the shooting position.The application can realize dynamic monitoring of the construction progress by determining the construction deviation, provide a data basis for the construction resource scheduling of the management personnel, reduce the work reporting threshold by determining the standardized work reporting audio, verify the construction deviation, effectively improve the accuracy of work reporting, and improve the digital and intelligent monitoring level of the bridge construction.

Description

Technical Field

[0001] This invention relates to the field of construction monitoring technology, and in particular to a method and system for intelligent construction monitoring of bridges. Background Technology

[0002] Bridge construction monitoring is a crucial aspect of ensuring the safety, efficiency, and quality of bridge construction. In current technologies, progress monitoring is generally carried out through manual inspections or data recording. However, this method not only increases operation and maintenance costs and burdens on construction site management, but also makes it difficult to form a complete database. This results in managers being unable to have an overall view of the construction progress, affecting adjustments to construction plans and resource allocation, and making it impossible to trace quality and progress.

[0003] If construction workers can take the initiative to upload construction information via mobile devices, the timeliness and completeness of the data can be greatly improved, forming a continuous and traceable time-series database. This source recording method not only greatly reduces data delays and information distortion, but also enhances data accuracy and avoids errors in manual transcription.

[0004] Therefore, how to use the construction workers' work reports in video for construction monitoring is the technical problem that this invention needs to solve. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for intelligent bridge construction monitoring, so as to solve the problem of how to use construction workers' work reports video for construction monitoring as mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for intelligent construction monitoring of bridges, the method comprising:

[0008] A panoramic image of the bridge construction site is drawn and divided into several partitions. Based on a preset numbering rule, a number is set for each partition. The construction tasks of the construction personnel are collected, and the corresponding partitions are defined as task areas. The monitoring period for bridge construction is delineated, and when the monitoring period arrives, a work reporting window is sent to the construction personnel, with work reporting rules embedded.

[0009] Read the work report video uploaded by the construction workers, identify the shooting location, and determine whether the shooting location is the same as the task area. If so, extract the key frame containing the task area from the work report video, extract the real-time image of the task area from the panoramic image, compare the key frame and the real-time image to obtain the daily progress, integrate the construction task and the daily progress, and generate the construction deviation.

[0010] Configure the priority of each construction task, define the risk level of construction deviation based on the priority, where each risk level corresponds to a handling rule, and activate the handling rule corresponding to the construction deviation;

[0011] Audio data is extracted from the work report video, key elements are identified, including at least the construction location, blind spot characteristics and status, and written into a preset template to generate standardized work report audio. The construction deviation is corrected, updated to the daily progress, a construction monitoring report is generated, and sent to a preset terminal.

[0012] Furthermore, the steps of drawing a panoramic image of the bridge construction site, dividing it into several partitions, assigning a number to each partition based on a preset numbering rule, and collecting construction tasks from the construction personnel include:

[0013] The causes of the construction deviations are identified through a process that is then used to establish a correspondence between the construction deviations and the zoning areas based on the construction tasks.

[0014] Insert tags generated by the cause into the partition, integrate all tags, generate a set of hindering factors, and send it to a preset terminal.

[0015] Furthermore, the step of collecting construction tasks from construction workers and defining the corresponding zones as task areas includes:

[0016] Collect the identity information of construction workers, integrate the identity information and the task area to obtain a task list, and grant modification permissions to the task list to preset terminals;

[0017] Embed scheduling relationships into the task list.

[0018] Furthermore, the step of reading the work report videos uploaded by construction workers and identifying the filming location includes:

[0019] Create a construction monitoring platform and upload the task list and panoramic images to the construction monitoring platform;

[0020] Select the ideal location from each task area and integrate it into the reporting window.

[0021] Furthermore, the step of extracting keyframes containing the task area includes:

[0022] From the keyframes, fuzzy features are selected, and monitoring data containing fuzzy features is collected using a preset monitoring device, and the fuzzy features are corrected.

[0023] The key elements are verified using the reported video and blurred features.

[0024] Furthermore, the step of configuring the priority of each construction task and defining the risk level of construction deviation based on the priority includes:

[0025] Record the number of days of construction deviation; when the number of days of deviation exceeds a threshold, adjust the risk level.

[0026] Introduce multi-factor indicators into the risk level, wherein the multi-factor indicators include at least: the complexity of the construction task and the natural environment.

[0027] Furthermore, the method also includes:

[0028] The monitoring equipment identifies potential hazards in each zone.

[0029] Based on the characteristics of construction deviations and potential hazards, a performance evaluation mechanism is constructed, evaluation results are generated, and published on the construction monitoring platform.

[0030] Furthermore, the method also includes:

[0031] Create a resource scheduling pool and identify idle resources in all partitions via the monitoring device, wherein the idle resources include at least: personnel and equipment;

[0032] Generate blocks that correspond one-to-one with the partitions. Based on the construction deviation, cluster the blocks into priority blocks and delay blocks, and link them to both sides of the resource scheduling pool respectively.

[0033] A tilt mechanism is embedded in the resource scheduling pool. When the resource scheduling pool tilts, the monitoring frequency of the monitoring devices in the priority block is adjusted. An invitation SMS is generated using the delay block and sent to the device terminal corresponding to the idle resource via the monitoring device.

[0034] Furthermore, the step of comparing the keyframes and real-time images to obtain the daily progress includes:

[0035] Create image matching models that correspond one-to-one with partitions, collect historical data of keyframes and real-time images, annotate them, generate training sets, train the image matching models, extract the model parameters of all image matching models, calculate the average parameters, and write them into the preset full-range model.

[0036] Keyframes and real-time images are input into the full-range model to obtain the daily progress. The image comparison model is activated at a preset frequency, and keyframes and real-time images are input into the image comparison model respectively to obtain the output results. It is determined whether the output results are the same as the daily progress. If not, comparison information is generated and sent to a preset terminal.

[0037] Furthermore, the system includes:

[0038] The embedded module is used to draw a panoramic image of the bridge construction site and divide it into several partitions. Based on the preset numbering rules, each partition is assigned a number. The construction tasks of the construction personnel are collected, and the corresponding partitions are defined as task areas. The monitoring period for bridge construction is defined, and when the monitoring period arrives, a work reporting window is sent to the construction personnel, and the work reporting rules are embedded.

[0039] The comparison module is used to read the work report video uploaded by the construction personnel, identify the shooting location, and determine whether the shooting location is the same as the task area. If so, it extracts the key frame containing the task area from the work report video, extracts the real-time image of the task area from the panoramic image, compares the key frame and the real-time image to obtain the daily progress, integrates the construction task and the daily progress, and generates the construction deviation.

[0040] The activation module is used to configure the priority of each construction task, define the risk level of construction deviation based on the priority, wherein each risk level corresponds to a handling rule, and activate the handling rule corresponding to the construction deviation.

[0041] The sending module is used to extract audio data from the work report video, identify key elements, wherein the key elements include at least: construction location, blind spot characteristics and status, and write them into a preset template to generate standardized work report audio, correct the construction deviation, update the daily progress, generate a construction monitoring report, and send it to a preset terminal.

[0042] Furthermore, the embedding module includes:

[0043] A unit is established to iterate through the causes of the construction deviations and, through the construction tasks, establish the correspondence between the construction deviations and the zones.

[0044] An insertion unit is used to insert tags generated by the cause into the partition, integrate all tags, generate a set of hindering factors, and send it to a preset terminal.

[0045] An open unit is used to collect the identity information of construction personnel, integrate the identity information and the task area to obtain a task list, and grant modification permissions to the task list to a preset terminal.

[0046] The scheduling unit is used to embed scheduling relationships into the task list.

[0047] Compared with the prior art, the beneficial effects of the present invention are:

[0048] By creating panoramic images, the construction site situation can be displayed intuitively, improving the efficiency of construction scheduling. By issuing reporting windows, the reporting threshold can be lowered, reducing the construction monitoring pressure on management personnel and facilitating the collection and analysis of construction data. By identifying construction deviations, dynamic monitoring of construction progress can be achieved, providing a data foundation for management personnel to allocate construction resources. By defining standardized reporting audio, the reporting threshold can be lowered, construction deviations can be verified, and the accuracy of reporting can be effectively improved, thus enhancing the digitalization and intelligence level of bridge construction monitoring. Attached Figure Description

[0049] Figure 1 A flowchart illustrating the intelligent construction monitoring method for bridges provided in this embodiment of the invention;

[0050] Figure 2 This is a first sub-flowchart of the intelligent bridge construction monitoring method provided in an embodiment of the present invention;

[0051] Figure 3 This is a second sub-flowchart of the intelligent bridge construction monitoring method provided in an embodiment of the present invention;

[0052] Figure 4 This is a third sub-flowchart of the intelligent bridge construction monitoring method provided in this embodiment of the invention;

[0053] Figure 5 This is a block diagram of the intelligent bridge construction monitoring system provided in an embodiment of the present invention;

[0054] Figure 6 This is a block diagram of the components of the embedded module in the intelligent bridge construction monitoring system provided in an embodiment of the present invention;

[0055] Figure 7 This is a block diagram of the comparison module in the intelligent bridge construction monitoring system provided in an embodiment of the present invention.

[0056] Figure 8 This is a block diagram of the activation module in the intelligent bridge construction monitoring system provided in an embodiment of the present invention. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0058] In Example 1, Figure 1 The implementation flow of the intelligent bridge construction monitoring method provided in this embodiment of the invention is illustrated below, and will be described in detail below:

[0059] S100: Draw a panoramic image of the bridge construction site and divide it into several partitions. Based on the preset numbering rules, set the number of each partition, collect the construction tasks of the construction personnel, define the corresponding partition as the task area, delineate the monitoring period of the bridge construction, and issue a work reporting window to the construction personnel when the monitoring period arrives, and embed the work reporting rules.

[0060] Using drones or panoramic cameras, image data of the bridge construction site is collected and a panoramic image is generated. According to the construction functional area, geographical location, or construction process, the panoramic image is divided into multiple zones using the grid method or polygon area method, such as the foundation construction area, main beam installation area, material stacking area, and temporary road area. Each zone is numbered using a preset numbering rule, where the numbering rule can be "block type code + sequence number", and the number is marked to the corresponding zone.

[0061] Collect the construction tasks of each construction worker. The construction tasks should include the task type, execution time, work content and work team. For example, the construction task received by a construction worker is: "Carry out mechanical excavation of bridge pier foundation in zone JZ-01. The planned start time is 8:00 on May 25 and it is expected to last for 2 days". The zone in which the construction worker performs the construction task is the task zone (i.e., zone JZ-01).

[0062] Based on the construction plan or the work schedule of the construction workers, the monitoring period for bridge construction is defined. The monitoring period is the time when the construction progress of the construction workers is collected. When the monitoring period arrives, a "reporting window" is issued to each construction worker. This window not only has an entry point for shooting the reporting video, but also embeds reporting rules, such as the construction progress to be described, the materials used, the completion rate, and the explanation of abnormal situations. The reporting rules can be displayed through voice broadcast.

[0063] For example, in the logistics and distribution field, after loading the goods, truck drivers usually take videos of the goods handover and loading completion on the booking platform. This is called a work report video, and the work report interface on the booking platform is the work report window.

[0064] S200: Read the work report video uploaded by the construction personnel, identify the shooting location, and determine whether the shooting location is the same as the task area. If so, extract the key frame containing the task area from the work report video, extract the real-time image of the task area from the panoramic image, compare the key frame and the real-time image to obtain the daily progress, integrate the construction task and the daily progress, and generate the construction deviation.

[0065] The system receives work reports uploaded by construction workers. By analyzing metadata (such as GPS location information) in the video or querying panoramic images using visual features, the system determines the video's shooting location. Using feature point recognition algorithms and deep learning models, it extracts environmental features from the video (such as building components, equipment markings, and ground markings) and compares them with the task area in the panoramic image. If the shooting location matches the worker's task area, keyframes containing the task area image are extracted from the video based on the environmental features. Real-time images of the task area are also extracted from the panoramic image, and the keyframes and real-time images are compared to analyze differences in structural form, material condition, and construction equipment. These differences are defined as daily progress. The system continues to compare daily progress information with pre-set construction tasks, considering task duration, process nodes, and construction quality standards to determine the deviation between daily progress and the construction task—the construction deviation. Construction deviation includes unfinished portions of the construction task, insufficient or excessive workload, etc.

[0066] S300: Configure the priority of each construction task, define the risk level of construction deviation based on the priority, wherein each risk level corresponds to a handling rule, and activate the handling rule corresponding to the construction deviation.

[0067] Based on factors such as the nature of the construction task, the criticality of the procedures, the difficulty of construction, and the impact on the overall progress, the priority of each construction task is determined. For example, key procedures such as foundation construction and installation of critical structures are set as high priority, while auxiliary operations are set as medium to low priority. The risk level of construction deviations is determined according to the priority of the construction tasks. For example, if a construction deviation corresponds to a construction task with a high priority, and the deviation is small, then the deviation is defined as a high-risk level. If another construction deviation corresponds to a construction task with a low priority, and the deviation is large, then the risk level of the deviation can be set as medium or high. Each risk level corresponds to a handling rule. For example, a construction task is the construction of temporary guardrails along the bridge. This task has a medium priority. If, by comparing the progress reports and panoramic images, a deviation is found between the daily progress and the construction task, and the deviation is assuming that 20 meters of temporary guardrails along the bridge remain to be installed, then the risk level of the deviation can be defined as high, and the deviation should be completed first on the next working day.

[0068] S400: Extract audio data from the work report video, identify key elements, wherein the key elements include at least: construction location, blind spot characteristics and status, and write them into a preset template to generate standardized work report audio, correct the construction deviation, update it to the daily progress, generate a construction monitoring report, and send it to a preset terminal.

[0069] Audio data was extracted from construction reports using audio-video separation technology. A speech recognition model was then used to convert the audio content into text. Natural language processing algorithms were combined to perform semantic analysis on the identified text, extracting key elements. These key elements included construction locations mentioned by workers (e.g., "pier No. 3," "main beam connection section"), blind spot features (e.g., "cracks found behind equipment," "material piled in corner"), on-site conditions (e.g., "water seepage," "loose rebar"), and construction progress (e.g., "surface painting mostly completed," "formwork erection completed"). These key elements were then used to extract the relevant information. The data is written into a preset template to generate standardized audio data for construction progress reporting in text format. Construction deviations are then corrected. The correction process includes: determining whether there are contradictory or complementary items between the standardized audio data and the construction deviations; if so, adjusting the deviations using complementary items and re-verifying the contradictory items on the next working day; generating a construction monitoring report using the corrected deviations, which can be simply understood as a summary of unfinished construction tasks each day; and sending the construction monitoring report to a preset terminal, which is also the bridge construction management personnel's terminal, for overall management by the management personnel.

[0070] In Example 2, Figure 2 The implementation flow of the intelligent bridge construction monitoring method provided by an embodiment of the present invention is illustrated. The following details the steps of drawing a panoramic image of the bridge construction site, dividing it into several partitions, setting a number for each partition based on a preset numbering rule, and collecting the construction tasks of the construction personnel:

[0071] S101: Traverse the causes of the construction deviations and establish the correspondence between the construction deviations and the zones through the construction tasks.

[0072] Identify the causes of each construction deviation, such as material supply delays, equipment failures, weather conditions, design changes, or misalignments in construction procedures; establish a correspondence between construction deviations and zones. This approach allows for the rapid identification of the corresponding construction personnel and facilitates progress tracking.

[0073] S102: Insert the tags generated by the cause into the partition, integrate all the tags, generate a set of obstructing factors, and send it to the preset terminal.

[0074] Insert labels generated from the causes into the partitions. These labels are similar to annotations and notes. Integrate all the labels (cause) to generate a set of hindering factors, which is the set of reasons that cause the expected progress of bridge construction to be hindered.

[0075] In Example 3, Figure 2The implementation flow of the intelligent bridge construction monitoring method provided by an embodiment of the present invention is illustrated. The following details the steps of collecting construction tasks from construction personnel and defining the corresponding partitions as task areas:

[0076] S103: Collect the identity information of construction personnel, integrate the identity information and task area to obtain a task list, and grant modification permissions to the task list to preset terminals.

[0077] The identity information (such as name and contact information) of each construction worker is determined through mobile device registration or manual reporting. Based on the allocation of construction tasks, the task area corresponding to each construction worker is determined, and a task list is generated. The task list consists of identity information items and task area items. The task list is stored and created by a preset terminal.

[0078] S104: Embed scheduling relationships into the task list.

[0079] Embed scheduling relationships into the task list. Scheduling relationships refer to the scheduling relationships between construction workers. For example, a scheduling relationship can be established between construction worker A and worker B. When it is confirmed through the work reporting video that worker A has completed the construction task, worker A can be assigned to the task area corresponding to worker B according to the scheduling relationship, and construction can continue.

[0080] In Example 4, Figure 3 The implementation flow of the intelligent bridge construction monitoring method provided in this embodiment of the invention is shown below. The steps of reading the work report video uploaded by the construction personnel and identifying the shooting location are described in detail below:

[0081] S201: Create a construction monitoring platform and upload the task list and panoramic image to the construction monitoring platform.

[0082] A construction monitoring platform was created, and task lists and panoramic images were uploaded to the platform. The construction monitoring platform is mainly used to manage and allocate construction tasks, and to dispatch personnel and materials. It can also provide an overview of the bridge construction progress.

[0083] S202: Select the ideal location from each task area and integrate it into the reporting window.

[0084] To lower the barrier to entry when shooting progress reports, ideal locations (i.e. ideal locations for shooting construction progress) can be entered into the progress report window to prompt construction workers to shoot from the designated locations; the ideal locations can then be integrated into the progress report window and sent to the construction workers.

[0085] In Example 5, Figure 3The implementation flow of the intelligent bridge construction monitoring method provided by the embodiment of the present invention is shown below. The step of extracting keyframes containing the task area is described in detail below:

[0086] S203: Select fuzzy features from the keyframes, collect monitoring data containing fuzzy features using a preset monitoring device, and correct the fuzzy features.

[0087] In keyframes, areas with blurred features are identified, and pre-set monitoring equipment (such as high-definition cameras, infrared sensors, and drones) is used to focus on collecting data from these areas to obtain higher-quality monitoring data. By comparing the data with clear features and combining data fusion or image enhancement processing algorithms, the collected blurred features are corrected, thereby improving the image details of the keyframes and enhancing the accuracy of construction monitoring.

[0088] S204: Verify the key elements using the reported video and blurred features.

[0089] By utilizing video reporting and fuzzy features, key elements are refined to enrich their details and improve their information accuracy.

[0090] In Example 6, Figure 4 The implementation flow of the intelligent bridge construction monitoring method provided by an embodiment of the present invention is illustrated. The following details the steps of configuring the priority of each construction task and defining the risk level of construction deviation based on the priority:

[0091] S301: Record the number of days of deviation for construction deviation. When the number of days of deviation exceeds the threshold, adjust the risk level.

[0092] Record the duration of construction deviations, i.e., the number of deviation days. The number of deviation days can also be understood as the actual number of days that each construction task deviates from the original construction plan. When the number of deviation days exceeds the threshold, the risk level corresponding to the construction deviation is increased.

[0093] S302: Introduce multi-factor indicators into the risk level, wherein the multi-factor indicators include at least: the complexity of the construction task and the natural environment.

[0094] In addition to being determined by construction deviations, risk levels are also related to multiple factors, such as the complexity of the construction task, the specific location of the construction, and weather and environment. If the construction task corresponding to a certain construction deviation is highly complex and the risk level of the construction deviation is medium, then the risk level can be increased to allocate more human and material resources.

[0095] In Example 7, unlike Example 1, the method further includes:

[0096] The monitoring equipment identifies potential hazards in each zone.

[0097] Based on the characteristics of construction deviations and potential hazards, a performance evaluation mechanism is constructed, evaluation results are generated, and published on the construction monitoring platform.

[0098] Using monitoring equipment, the characteristics of potential hazards in each zone are collected. These hazards can include surface water accumulation, lack of warning signs, etc. Weights and scoring standards are set for each type of construction deviation and hazard to build a performance evaluation mechanism. The performance of construction personnel at the bridge construction site is regularly evaluated, and evaluation results reports are generated and published on the construction monitoring platform. The results are presented in the form of visual charts or alarm signs to facilitate dynamic supervision and progress tracking by management personnel.

[0099] In Example 8, unlike Example 1, the method further includes:

[0100] Create a resource scheduling pool and identify idle resources in all partitions via the monitoring device, wherein the idle resources include at least: personnel and equipment;

[0101] Generate blocks that correspond one-to-one with the partitions. Based on the construction deviation, cluster the blocks into priority blocks and delay blocks, and link them to both sides of the resource scheduling pool respectively.

[0102] A tilt mechanism is embedded in the resource scheduling pool. When the resource scheduling pool tilts, the monitoring frequency of the monitoring devices in the priority block is adjusted. An invitation SMS is generated using the delay block and sent to the device terminal corresponding to the idle resource via the monitoring device.

[0103] A resource scheduling pool is created, which is a logical representation of the resource management and scheduling platform, similar to a parent node in a binary tree in existing technologies. Monitoring equipment (such as high-definition cameras, infrared sensors, and drones) is used to determine whether there are idle resources in the partitions. Idle resources refer to personnel and equipment that are not in use or working. Blocks are created corresponding to each partition, similar to child nodes in a binary tree. Blocks are divided into priority blocks and delayed blocks. Priority blocks are those with higher progress, while delayed blocks are those with relatively lower progress. Priority blocks and delayed blocks are linked to the two sides of the resource scheduling pool to form a dynamically adjustable resource management structure. The linking process is similar to attaching a child node to the parent node corresponding to the resource scheduling pool. A tilting mechanism is embedded in the resource scheduling pool. The tilting mechanism is as follows: when the construction deviation of priority blocks and delayed blocks exceeds a preset fluctuation range, the horizontal angle of the resource scheduling pool is tilted. The tilting process is similar to rotating the binary tree vertically around the parent node.

[0104] When the resource scheduling pool tilts, it indicates that the advance of the priority block or the sluggish progress of the delay block exceeds the preset fluctuation range. The construction progress needs to be dynamically adjusted immediately. The adjustment process includes: increasing the monitoring frequency of the monitoring equipment in the priority block to obtain more idle resources, and generating an invitation SMS using the location and specific progress of the delay block; controlling the monitoring equipment to restart, and using the Bluetooth device pre-integrated in the monitoring equipment to establish a data channel with the user terminal of the idle resource, and sending the invitation SMS to the user terminal through this data channel to remind the user to go to the delay block to participate in the construction.

[0105] In Example 9, unlike Example 1, the step of comparing the keyframes and real-time images to obtain the daily progress includes:

[0106] Create image matching models that correspond one-to-one with partitions, collect historical data of keyframes and real-time images, annotate them, generate training sets, train the image matching models, extract the model parameters of all image matching models, calculate the average parameters, and write them into the preset full-range model.

[0107] Keyframes and real-time images are input into the full-range model to obtain the daily progress. The image comparison model is activated at a preset frequency, and keyframes and real-time images are input into the image comparison model respectively to obtain the output results. It is determined whether the output results are the same as the daily progress. If not, comparison information is generated and sent to a preset terminal.

[0108] Based on deep learning algorithms, an image comparison model is created, where each partition corresponds to an image comparison model. Historical data of keyframes and real-time images are collected, and professionals annotate the historical data, such as the location, category, change type, and abnormal state of the target object. A training set is constructed using the annotated data, with one training set for each partition. The image comparison model for each partition is trained using the training set of each partition. The model parameters of all image comparison models are extracted, including convolution kernel weights, bias terms, feature extraction layer weights, and normalization layer parameters. A weight value is set for each model parameter, and the average parameter is calculated by weighted averaging. All partitions correspond to a set of average parameters, which are written into the initialized image comparison model to obtain the full-range model; in other words, the image comparison model corresponding to all partitions is the full-range model.

[0109] During routine monitoring, keyframes and real-time images are input into the full-range model to output daily progress. It's important to note that during daily monitoring, the output of the full-range model is directly used as the daily progress. At a preset frequency (daily or every few days), partitions with progress discrepancies are activated, and the corresponding keyframes and real-time images are input into the image comparison model to obtain the output. The image comparison model is then used to cross-validate the output of the full-range model. If they match, the matching result is used as the daily progress; otherwise, the outputs of both the image comparison model and the full-range model are written into a preset template to generate a comparison message, which is then sent to a preset terminal for manual evaluation by bridge construction management personnel.

[0110] In this embodiment, by constructing a full-range model, unified analysis and judgment of daily monitoring data can be achieved, and the daily progress difference of each zone can be quickly identified. At the same time, within a preset time window, the image comparison model of the zone corresponding to the progress difference can be retrieved and activated to perform cross-validation and accuracy verification of the progress difference, thereby improving the accuracy of progress difference judgment.

[0111] Figure 5 This diagram illustrates the structural block diagram of a bridge intelligent construction monitoring system provided in an embodiment of the present invention. The bridge intelligent construction monitoring system 1 includes:

[0112] The embedded module 11 is used to draw a panoramic image of the bridge construction site and divide it into several partitions. Based on the preset numbering rules, each partition is assigned a number. The construction tasks of the construction personnel are collected, and the corresponding partition is defined as the task area. The monitoring period of the bridge construction is defined, and when the monitoring period arrives, a reporting window is sent to the construction personnel, and the reporting rules are embedded.

[0113] The comparison module 12 is used to read the work report video uploaded by the construction personnel, identify the shooting location, and determine whether the shooting location is the same as the task area. If so, it extracts the key frame containing the task area from the work report video, extracts the real-time image of the task area from the panoramic image, compares the key frame and the real-time image to obtain the daily progress, integrates the construction task and the daily progress, and generates the construction deviation.

[0114] The activation module 13 is used to configure the priority of each construction task, define the risk level of construction deviation based on the priority, wherein each risk level corresponds to a handling rule, and activate the handling rule corresponding to the construction deviation.

[0115] The sending module 14 is used to extract audio data from the work report video, identify key elements, wherein the key elements include at least: construction location, blind spot characteristics and status, and write them into a preset template to generate work report audio, correct the construction deviation, update it to the daily progress, generate a construction monitoring report, and send it to a preset terminal.

[0116] Figure 6 This diagram illustrates the structural block diagram of the intelligent bridge construction monitoring system provided in an embodiment of the present invention. The embedded module 11 includes:

[0117] Establishment unit 111 is used to traverse the causes of the construction deviations and establish the correspondence between the construction deviations and the zones through the construction tasks;

[0118] Insertion unit 112 is used to insert tags generated by the cause into the partition, integrate all tags, generate a set of obstructing factors, and send it to a preset terminal;

[0119] The open unit 113 is used to collect the identity information of construction personnel, integrate the identity information and the task area to obtain a task list, and grant modification permissions to the task list to a preset terminal.

[0120] The scheduling unit 114 is used to embed scheduling relationships into the task list.

[0121] Figure 7 This diagram illustrates the structural composition of the intelligent bridge construction monitoring system provided in an embodiment of the present invention. The comparison module 12 includes:

[0122] Upload unit 121 is used to create a construction monitoring platform and upload the task list and panoramic image to the construction monitoring platform;

[0123] Integration unit 122 is used to select an ideal location from each task area and integrate it into the reporting window;

[0124] The correction unit 123 is used to select fuzzy features from the key frame, collect monitoring data containing fuzzy features using a preset monitoring device, and correct the fuzzy features.

[0125] Verification unit 124 is used to verify the key elements using the reported video and blurry features.

[0126] Figure 8 This diagram illustrates the structural composition of the intelligent bridge construction monitoring system provided in an embodiment of the present invention. The activation module 13 includes:

[0127] The adjustment unit 131 is used to record the number of days of deviation for construction deviation, and adjust the risk level when the number of days of deviation is greater than a threshold.

[0128] The introduction unit 132 is used to introduce multi-factor indicators into the risk level, wherein the multi-factor indicators include at least: the complexity of the construction task and the natural environment.

[0129] The embedding module 11 is mainly used to complete step S100, the comparison module 12 is mainly used to complete step S200, the activation module 13 is mainly used to complete step S300, and the sending module 14 is mainly used to complete step S400.

[0130] The establishment unit 111 is mainly used to complete step S101, the insertion unit 112 is mainly used to complete step S102, the opening unit 113 is mainly used to complete step S103, and the scheduling unit 114 is mainly used to complete step S104.

[0131] Upload unit 121 is mainly used to complete step S201, integration unit 122 is mainly used to complete step S202, correction unit 123 is mainly used to complete step S203, and verification unit 124 is mainly used to complete step S204.

[0132] The adjustment unit 131 is mainly used to complete step S301, and the introduction unit 132 is mainly used to complete step S302.

[0133] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0134] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0135] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A bridge intelligent construction monitoring method, characterized in that, The method comprises: drawing a panoramic image of a bridge construction site, and cutting it into a plurality of partitions, setting a number of each partition based on a preset numbering rule, collecting construction tasks of construction personnel, and defining a corresponding partition as a task area, delimiting a monitoring period of the bridge construction, and issuing a work reporting window to the construction personnel when the monitoring period arrives, and embedding a work reporting rule; reading a work reporting video uploaded by the construction personnel, identifying a shooting position, judging whether the shooting position is the same as the task area, if so, intercepting a key frame containing the task area from the work reporting video, extracting a real-time image of the task area from the panoramic image, comparing the key frame and the real-time image to obtain a daily progress, integrating the construction task and the daily progress to generate a construction deviation; configuring a priority of each construction task, defining a risk level of the construction deviation based on the priority, wherein each risk level corresponds to a disposal rule, and activating the disposal rule corresponding to the construction deviation; extracting audio data in the work reporting video, identifying key elements, wherein the key elements at least include: a construction site, a blind area feature and a state, and writing into a preset template to generate a standardized work reporting audio, correcting the construction deviation, updating to the daily progress, generating a construction monitoring report, and sending to a preset terminal; the step of reading the work reporting video uploaded by the construction personnel and identifying the shooting position comprises: creating a construction monitoring platform, uploading a task list and a panoramic image to the construction monitoring platform; selecting an ideal position from each task area and integrating it into the work reporting window; the step of comparing the key frame and the real-time image to obtain the daily progress comprises: creating an image comparison model corresponding to each partition, collecting historical data of the key frame and the real-time image, labeling, generating a training set, training the image comparison model, extracting model parameters of all image comparison models, calculating average parameters, and writing into a preset full-range model; inputting the key frame and the real-time image into the full-range model to obtain the daily progress, activating the image comparison model according to a preset frequency, inputting the key frame and the real-time image into the image comparison model respectively to obtain an output result, judging whether the output result is the same as the daily progress, if not, generating comparison information and sending to a preset terminal.

2. The bridge intelligent construction monitoring method of claim 1, wherein, the step of drawing a panoramic image of a bridge construction site, and cutting it into a plurality of partitions, setting a number of each partition based on a preset numbering rule, collecting construction tasks of construction personnel comprises: traversing the causes of the construction deviation, establishing a correspondence between the construction deviation and the partition through the construction task; inserting a label generated from the cause into the partition, integrating all labels to generate a set of blocking factors, and sending to a preset terminal.

3. The bridge intelligent construction monitoring method of claim 1, wherein, the step of collecting construction tasks of construction personnel and defining a corresponding partition as a task area comprises: collecting identity information of construction personnel, integrating the identity information and the task area to obtain a task list, opening modification authority of the task list to a preset terminal; embedding a scheduling relationship in the task list.

4. The bridge intelligent construction monitoring method of claim 1, wherein, The step of intercepting the key frame containing the task area comprises: From the key frame, select the fuzzy feature, collect the monitoring data containing the fuzzy feature by using the preset monitoring device, and correct the fuzzy feature; Verify the key elements by using the reporting video and the fuzzy feature.

5. The bridge intelligent construction monitoring method of claim 1, wherein, The step of configuring the priority of each construction task, and defining the risk level of the construction deviation based on the priority comprises: Record the deviation days of the construction deviation, and adjust the risk level when the deviation days are greater than a threshold value; Introduce a multi-factor index into the risk level, wherein the multi-factor index at least includes the complexity of the construction task and the natural environment.

6. The bridge intelligent construction monitoring method of claim 4, wherein, The method further comprises: Identify the hidden danger feature in each subarea via the monitoring device; Based on the construction deviation and the hidden danger feature, construct a performance evaluation mechanism, generate an evaluation result, and publish it to the construction monitoring platform.

7. The bridge intelligent construction monitoring method of claim 6, wherein, The method further comprises: Create a resource scheduling pool, identify the idle resources in all subareas via the monitoring device, wherein the idle resources at least include personnel and equipment; Generate a block corresponding to each subarea, cluster the blocks into priority blocks and delay blocks based on the construction deviation, and link them to the two sides of the resource scheduling pool respectively; Embed an inclination mechanism into the resource scheduling pool, adjust the monitoring frequency of the monitoring device in the priority block when the resource scheduling pool is inclined, generate an invitation message by using the delay block, and send it to the device terminal corresponding to the idle resources via the monitoring device.

Images