Component difference detection method, device and equipment based on process constraint and medium

Abstract

This invention relates to the field of building information processing technology, and provides a method, device, equipment, and medium for component difference detection based on process constraints. It can construct a 4D process sequence constraint model based on BIM models, construction schedules, and work breakdown structure process data, and for the first time transforms process sequence logic into a computable constraint model. Based on pose information and visibility confidence, multiple BIM candidate components are selected from the theoretically expected component set, and matching relationships are established through category priors and spatial proximity. Multi-level filtering significantly narrows the matching search space, and the matching relationships establish a bridge between on-site observed components and BIM components. Using the 4D process sequence constraint model as a constraint for difference detection effectively reduces the false alarm rate and false negative rate, and the difference judgment results are closely related to process and time, providing a clear basis for subsequent explanation and rectification.

Description

Technical Field

[0001] This invention relates to the field of building information processing technology, and in particular to a method, apparatus, equipment and medium for detecting component differences based on process constraints. Background Technology

[0002] In recent years, the integration of BIM (Building Information Modeling) with construction site images (such as videos captured by fixed cameras, mobile terminals, or drones) has been widely used for construction progress verification, quality inspection, and safety management. Specifically, existing technologies typically employ the following solutions: 1. Extract "constructed" components or component outlines from on-site images using computer vision algorithms (such as object detection, semantic segmentation, point cloud reconstruction, etc.); 2. Project the corresponding components in the BIM model to the same viewpoint or spatial coordinates; 3. Determine whether the model is consistent with the actual site based on geometric overlap, appearance similarity, voxel occupancy, and other methods.

[0003] However, the above solution has obvious flaws: 1. Lack of process sequence constraints leads to many false alarms and omissions: There are a large number of temporary components, material piles, scaffolding and obstructions at the construction site. If only static comparison is made, it is easy to misjudge the "missing components that have not yet been planned for construction" as a difference, and to miss the situation of "unplanned early construction". 2. Large search space and low efficiency: Traditional methods require matching images and observing them one by one in all BIM components, which is computationally intensive and difficult to apply in real time; 3. Difficulty in handling phased occlusion and uncertainties in perspective during construction: The construction process is highly dynamic, and the visibility of components changes with the sequence of procedures; without considering the sequence of procedures, the system has difficulty distinguishing between "occlusion causing invisibility" and "actual missing parts"; 4. Progress or quality verification relies on manual review: Even if discrepancies are detected, there is a lack of corresponding explanations in the 4D construction plan, making it difficult to automatically generate executable rectification and progress feedback. Summary of the Invention

[0004] In view of the above, it is necessary to provide a component difference detection method, device, equipment and medium based on process constraints, in order to solve the problems of low accuracy and lack of interpretability of existing component difference detection methods.

[0005] A component difference detection method based on process constraints, the component difference detection method based on process constraints includes: In response to component difference detection commands, acquire BIM model, construction schedule and work breakdown structure process data, and current construction site image sequence; A 4D process timing constraint model is constructed based on the BIM model, the construction schedule, and the work breakdown structure process data. The theoretical set of components that should appear and the theoretical set of components that should not appear at each time point are then constructed based on the 4D process timing constraint model. Acquire target scene images and acquire the pose information of the camera corresponding to the current moment image; Based on the pose information, each BIM component in the theoretically expected component set at the current moment is projected onto the target site image to obtain the BIM component view projection, and the expected observation area and visibility confidence of each BIM component are generated based on the BIM component view projection. Multiple on-site observation components were obtained by performing instance recognition on the target site image; Based on the pose information and the visibility confidence, multiple BIM candidate components are selected from each BIM component in the theoretically expected component set, and a matching relationship is established between each on-site observed component and each BIM candidate component. Using the 4D process sequence constraint model as a constraint, difference detection is performed based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observation components to obtain the target difference detection result.

[0006] A component difference detection device based on process constraints, the component difference detection device based on process constraints includes: The acquisition unit is used to acquire the BIM model, construction schedule and work breakdown structure process data, and the current construction site image sequence in response to the component difference detection command. The construction unit is used to construct a 4D process timing constraint model based on the BIM model, the construction schedule plan and the work breakdown structure process data, and to construct the theoretically expected component set and the theoretically unexpected component set at each time point based on the 4D process timing constraint model. The acquisition unit is also used to acquire target scene images and acquire the pose information of the camera corresponding to the current moment image; The projection unit is used to project each BIM component in the theoretically expected component set at the current moment onto the target site image according to the pose information to obtain the BIM component view projection, and generate the expected observation area and visibility confidence of each BIM component according to the BIM component view projection. The identification unit is used to perform instance identification on the target scene image to obtain multiple scene observation components; The filtering unit is used to filter multiple BIM candidate components from each BIM component in the theoretically expected component set based on the pose information and the visibility confidence, and to establish a matching relationship between each on-site observed component and each BIM candidate component. The detection unit is used to perform difference detection based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observation components, using the 4D process sequence constraint model as a constraint, to obtain the target difference detection result.

[0007] A computer device, the computer device comprising: A memory for storing at least one instruction; and a processor for executing the instructions stored in the memory to implement the component difference detection method based on process constraints.

[0008] A computer-readable storage medium storing at least one instruction, which is executed by a processor in a computer device to implement the component difference detection method based on process constraints.

[0009] As can be seen from the above technical solutions, this invention can construct a 4D process sequence constraint model based on BIM model, construction schedule plan, and work breakdown structure process data, and for the first time transforms process sequence logic into a computable constraint model; based on pose information and visibility confidence, multiple BIM candidate components are selected from each BIM component in the theoretically expected component set, and a matching relationship is established between each on-site observed component and each BIM candidate component. Through multi-level filtering, the matching search space is significantly reduced, and the subsequent matching efficiency is significantly improved. Moreover, the matching relationship establishes a connection bridge between on-site observed components and BIM components, laying the foundation for difference determination; using the 4D process sequence constraint model as a constraint, difference detection is performed based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and multiple on-site observed components, effectively reducing the false alarm rate and false negative rate. Furthermore, the difference determination results are closely related to the process and time, providing a clear basis for subsequent explanation and rectification. Attached Figure Description

[0010] Figure 1 This is a flowchart of a preferred embodiment of the component difference detection method based on process constraints of the present invention; Figure 2 This is a functional block diagram of a preferred embodiment of the component difference detection device based on process constraints of the present invention; Figure 3 This is a schematic diagram of the structure of a computer device that implements a preferred embodiment of the component difference detection method based on process constraints according to the present invention. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0012] like Figure 1 The diagram shown is a flowchart of a preferred embodiment of the component difference detection method based on process constraints of the present invention. The order of the steps in this flowchart can be changed, and some steps can be omitted, depending on different requirements.

[0013] The component difference detection method based on process constraints is applied to one or more computer devices. The computer device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0014] The computer device can be any electronic product that can interact with the user, such as a personal computer, tablet computer, smartphone, personal digital assistant (PDA), game console, interactive network television (IPTV), smart wearable device, etc.

[0015] The computer equipment may also include network equipment and / or user equipment. The network equipment includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of hosts or network servers.

[0016] The server can be a standalone server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0017] Artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0018] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0019] The network in which the computer device is located includes, but is not limited to, the Internet, wide area network, metropolitan area network, local area network, and virtual private network (VPN).

[0020] S10, in response to the component difference detection command, acquires the BIM model (Building Information Modeling), construction schedule and work breakdown structure (WBS) process data, and the current construction site image sequence.

[0021] In this embodiment, the component difference detection command can be triggered by relevant personnel according to actual detection needs.

[0022] In this embodiment, the BIM model may include, but is not limited to, the following data: (1) Component geometric information: stored in the form of mesh or parametric volume, used to accurately describe the three-dimensional shape of the component; (2) Component Unique Identifier ID: Assign a unique code to each component to ensure the uniqueness of subsequent matching and traceability; (3) Component category (such as beam, column, slab, wall, etc.), spatial location coordinates (X / Y / Z three-dimensional coordinates) and associated floor or axis information, used to clarify the engineering attributes and spatial ownership of the component.

[0023] In this embodiment, the work breakdown structure process data may include, but is not limited to, the following data: (1) The planned start time and planned end time of each process; (2) The binding relationship between process and component (one process can correspond to multiple components, such as "first floor column casting process" corresponding to all column components on the first floor). (3) Constraints between processes, including prerequisite relationships (such as "formwork" being a prerequisite process for "pouring"), subsequent relationships, parallel relationships (such as "building the east unit of the first floor" and "building the west unit of the first floor"), and conflict relationships (such as "reinforcing bar binding" and "pipeline installation" in the same area cannot be carried out at the same time).

[0024] In this embodiment, the construction site image sequence may include image sequences such as fixed camera monitoring screens, photos taken by mobile terminals, and video frames from drone inspections.

[0025] In this embodiment, the timestamp of each image or video frame is extracted and unified with the construction plan time into the same engineering clock system, thereby eliminating the deviation in the time dimension.

[0026] Through the above embodiments, a basic data system for difference detection was constructed to ensure that the BIM model, process plan, and on-site images are complete and correlated in terms of data dimensions; time synchronization also provides a unified time reference for subsequent 4D temporal constraint modeling, avoiding constraint judgment errors caused by time deviation.

[0027] S11. Construct a 4D process timing constraint model based on the BIM model, the construction schedule, and the work breakdown structure process data, and construct the theoretically expected component set and the theoretically unexpected component set for each time step based on the 4D process timing constraint model.

[0028] In this embodiment, the step of constructing a 4D process sequence constraint model based on the BIM model, the construction schedule, and the work breakdown structure process data includes: Extract each process and the relationships between every two processes from the construction schedule and the work breakdown structure process data; wherein, the relationships between every two processes include preceding relationships, subsequent relationships, parallel relationships, and conflict relationships; Using each process as a node, and constructing edges based on the relationship between every two processes, a timing constraint graph is obtained; The planned time interval for each process is determined based on the construction schedule and the work breakdown structure process data. Each BIM component is extracted from the BIM model, and its status is mapped according to the planned time interval of each process; wherein, the process status of each BIM component includes not under construction, under construction, and under construction. The state of each BIM component is added to the timing constraint diagram as an attribute association object of each node to obtain the 4D process timing constraint model.

[0029] The associated attributes of each node can also include process ID, planned time range, and bound components.

[0030] Among them, "not under construction" indicates that the component binding process has not reached the planned start time; "under construction" indicates that the component binding process is between the planned start time and end time; and "under construction" indicates that the component binding process has exceeded the planned end time.

[0031] In the above embodiments, the process timing logic is transformed into a computable constraint model for the first time, providing a time-dimensional screening basis for subsequent difference detection.

[0032] In this embodiment, constructing the theoretically expected set of components and the theoretically unexpected set of components at each moment based on the 4D process timing constraint model includes: For each BIM component at any given time, the planned time interval of the corresponding process of the BIM component is obtained according to the 4D process sequence constraint model, and the buffer window is obtained. Add the buffer window to the planned time interval to obtain a buffer space, and check whether the moment is within the buffer space; When the moment is detected to be within the buffer zone, the timing constraint diagram is used to detect whether the preceding process of the process corresponding to the BIM component has been completed or is in a parallel state. When it is detected that the preceding process of the process corresponding to the BIM component has been completed or is in a parallel state, the BIM component is added to the theoretically expected component set; or When it is detected that the preceding process of the process corresponding to the BIM component is not completed and is not in a parallel state, the BIM component is added to the theoretically non-existent component set.

[0033] The size of the buffer window can be adjusted according to the complexity of the project to handle minor planning deviations (such as a 12-hour delay in process start-up due to material delays).

[0034] The theory should include a component set for storing components whose binding process has started at the corresponding time, whose preconditions are met, and which are within the time buffer range.

[0035] The theory should not include a component set used to store components whose binding process has not started, whose preconditions have not been met, or whose time buffer has been exceeded at the corresponding time.

[0036] In the above embodiments, the dynamically generated component set can significantly narrow the search range for subsequent matching, reducing the amount of computation from the source.

[0037] S12, acquire the target scene image, and acquire the pose information of the camera corresponding to the current moment image.

[0038] In this embodiment, the target site images can be collected using various sensor sources such as drones and fixed cameras.

[0039] In this embodiment, distortion correction, noise reduction, and brightness equalization can be performed sequentially on the construction site image sequence to obtain the target site image.

[0040] In the process of distortion correction, camera intrinsic parameters (such as focal length, principal point coordinates, distortion coefficients, etc.) can be used to correct radial and tangential distortion of the construction site image sequence, thereby eliminating imaging deviations caused by camera hardware.

[0041] In the process of denoising, Gaussian filtering and median filtering algorithms can be used to remove salt-and-pepper noise and Gaussian noise (such as blur noise caused by airflow turbulence during drone shooting).

[0042] In particular, when performing brightness equalization, histogram equalization or adaptive brightness adjustment can be used to solve the problem of uneven image brightness caused by backlighting and shadows, ensuring clear component features.

[0043] In the above embodiments, preprocessing can eliminate interference factors and provide high-quality input data for component identification.

[0044] In this embodiment, the camera's extrinsic pose (including the camera's three-dimensional position and three-dimensional attitude angle) can be estimated using methods such as SLAM (Simultaneous Localization and Mapping), multi-view geometry, or known control points (such as GPS (Global Positioning System) positioning stakes pre-set at the construction site).

[0045] In the above embodiments, the mapping relationship between the image and the three-dimensional space was established by camera pose estimation, laying the foundation for the projection of BIM components onto the image plane.

[0046] Of course, for PTZ cameras deployed at fixed locations with preset intrinsic and extrinsic parameters, their pose is known static data, so there is no need to estimate the camera pose in real time.

[0047] In this embodiment, if the image is a video or a multi-view sequence, a local point cloud (a set of points composed of a large number of three-dimensional points) or a depth map can be reconstructed based on the estimated pose information, thereby supplementing the spatial depth information and enhancing the stability of subsequent spatial matching.

[0048] In the above embodiments, the spatial information dimension is enhanced by reconstructing local point clouds or depth maps, thereby improving the robustness of subsequent matching.

[0049] S13, based on the pose information, project each BIM component in the theoretically expected component set at the current moment onto the target site image to obtain the BIM component view projection, and generate the expected observation area and visibility confidence of each BIM component based on the BIM component view projection.

[0050] In this embodiment, the camera extrinsic pose and imaging model in the pose information can be used to project each BIM component in the theoretically expected component set at the current moment onto a two-dimensional image plane to obtain the theoretical position of the component in the image.

[0051] In this embodiment, generating the expected observation area and visibility confidence level of each BIM component based on the view projection of the BIM component includes: Based on the view projection of the BIM component, generate the contour mask, bounding box and visible face of each BIM component, and generate the desired observation area of ​​each BIM component based on the contour mask, bounding box and visible face of each BIM component. For each BIM component, extract the exposed surface of the BIM component facing the current camera, and extract a preset number of sampling points from the exposed surface; Starting from the center of the camera, ray detection is performed along the line of sight at each sampling point to determine whether each ray intersects with other components before reaching the corresponding sampling point; The sampling points corresponding to rays that intersect with other components are determined as geometric occlusion sampling points, and the sampling points corresponding to rays that do not intersect with other components are determined as geometrically visible sampling points. The geometric visibility rate of the BIM component is obtained by calculating the quotient of the number of geometrically visible sampling points and the preset number. Identify each occlusion component involved in the occlusion, and determine the process state of each occlusion component according to the 4D process timing constraint model; The occlusion rule library is retrieved, and the visibility adjustment factor for each occlusion component is obtained by traversing the occlusion rule library based on the process status of each occlusion component; wherein, the occlusion rule library is used to store the correspondence between component type, process status, and visibility adjustment factor; The minimum value of the visibility adjustment factor of all occluded components is obtained as the target visibility adjustment factor of the BIM component; The visibility confidence level of the BIM component is obtained by multiplying the geometric visibility rate of the BIM component with the target visibility adjustment factor.

[0052] The contour mask is a binary image used to precisely outline the edges of the component.

[0053] The bounding box is a rectangular frame that surrounds the component.

[0054] The visible sheet refers to the surface area of ​​the component that can be observed in the image.

[0055] The preset number can be configured according to actual needs, such as 50 points.

[0056] The visibility adjustment factor for each occlusion component can be directly queried through the occlusion rule library. For example, if the occlusion component is an auxiliary facility permitted by the current process, such as formwork or scaffolding, and its process status is under construction or already under construction, it can be determined as a legal occlusion with a visibility adjustment factor of 0.9. If the occlusion component is an unplanned component, such as a component that has not yet reached its construction time or temporarily stacked materials, and its process status is not under construction, it can be determined as an abnormal occlusion with a visibility adjustment factor of 0.3.

[0057] In the above embodiments, the generated expected observation area provides a theoretical benchmark for subsequent comparison with on-site observations. Visibility confidence also quantifies the observability of components, avoiding misjudging invisibility due to occlusion as component loss, and providing a weighted basis for subsequent difference determination.

[0058] S14, perform instance recognition on the target site image to obtain multiple site observation components.

[0059] In this embodiment, target detection algorithms or instance segmentation algorithms can be used to analyze the preprocessed target construction site image in order to identify and segment the construction components in the image.

[0060] The multiple field observation components may include the core information of each identified component: category (corresponding to the BIM component category, such as "reinforced concrete column"), pixel region (such as contour mask or bounding box coordinates), identification confidence (i.e., the credibility of the corresponding algorithm for the identification result, with a value range of 0-1) and estimated spatial location (such as three-dimensional coordinates inferred from camera pose and pixel coordinates).

[0061] In the above embodiments, the deep network algorithm was used to achieve accurate identification and information extraction of on-site components, providing on-site observation data for comparison with the BIM model.

[0062] S15, based on the pose information and the visibility confidence level, select multiple BIM candidate components from each BIM component in the theoretically expected component set, and establish a matching relationship between each on-site observed component and each BIM candidate component.

[0063] In this embodiment, the step of selecting multiple BIM candidate components from each BIM component in the theoretically expected component set based on the pose information and the visibility confidence level includes: Based on the camera extrinsic parameters and view frustum parameters in the pose information, each BIM component is projected to the current viewpoint to obtain the projection result. Based on the projection result, components that do not intersect with the view frustum or whose projected area is smaller than a preset area are removed from each BIM component to obtain a subset of components that should appear in the field of view. BIM components with visibility confidence scores lower than the confidence threshold are removed from the subset of components that should appear within the field of view to obtain the plurality of BIM candidate components.

[0064] The preset area and the confidence threshold can be configured according to the actual needs of the scenario.

[0065] For example, based on the camera extrinsic parameters and view frustum parameters (the cone-shaped area within the camera's field of view) in the pose information, components that should theoretically appear in the component set can be projected onto the current viewpoint, and components that do not intersect with the view frustum or whose projected area is less than 50 pixels (a customizable threshold) can be removed to obtain a subset of components that should appear within the view domain. Further, by removing BIM components with visibility confidence scores lower than the confidence threshold from the subset of components that should appear within the view domain, the multiple BIM candidate components can be obtained.

[0066] In the above embodiments, multi-level filtering significantly reduces the matching search space, filtering all BIM components (e.g., thousands) into a small number of candidate components (e.g., dozens), which significantly improves the efficiency of subsequent matching.

[0067] In this embodiment, an initial matching relationship is established between the on-site observed component and the BIM candidate component by using category prior (e.g., the "reinforced concrete column" in image recognition only matches column components in BIM) and spatial proximity relationship (e.g., the spatial distance between the three-dimensional coordinates of the on-site observed component and the spatial position of the BIM candidate component is less than 0.5m), thus obtaining the matching relationship.

[0068] Through the above embodiments, a bridge can be established between on-site observation components and BIM components, laying the foundation for difference determination.

[0069] S16, using the 4D process sequence constraint model as a constraint, perform difference detection based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observation components to obtain the target difference detection result.

[0070] In this embodiment, the step of using the 4D process sequence constraint model as a constraint, and performing difference detection based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observed components to obtain the target difference detection result includes: For each candidate BIM component, if the overlap between the candidate BIM component and all field observation components is less than the overlap threshold or no field observation component matching the candidate BIM component is found in the matching relationship, it is determined that there is a missing difference. For each field observation component, if the field observation component belongs to the set of components that should not appear according to theory, and the identification confidence of the field observation component is greater than a preset threshold, it is determined that there is a difference between early appearance and unplanned appearance. According to the matching relationship, when it is detected that the overlap between a candidate BIM component and the matching field observation component is greater than or equal to the overlap threshold, and the displacement is greater than a preset displacement, the scale difference is greater than the scale difference threshold, and the morphological difference is greater than a preset amount, it is determined that there is a geometric or positional deviation. For each field observation component, if the field observation component belongs to the theoretically expected component set, the visibility confidence of the field observation component is less than the confidence threshold, and no candidate BIM component matching the field observation component is found in the matching relationship, it is determined that there is an occlusion uncertainty difference, and the alarm priority of the field observation component is reduced. The detected differences are adjusted using the 4D process timing constraint model as a constraint to obtain the target difference detection result.

[0071] The overlap threshold, the preset threshold, the preset displacement, the scale difference threshold, and the preset amount can be optimal values ​​selected based on a large number of implementations.

[0072] For example: If a candidate BIM component belongs to the theoretically expected component set, but its overlap with all field-observed components (i.e., the intersection or union area of ​​the expected observation area and the field observation area) is <0.3 (a customizable threshold) or there is no matching field-observed component, then a missing difference is determined. If a field-observed component belongs to the theoretically not expected component set, but the recognition confidence of the corresponding field image is >0.8 (a customizable threshold), then a difference between early appearance and unplanned appearance is determined. If the overlap between a candidate BIM component and a matching field-observed component is greater than or equal to 0.3, but there is significant displacement (e.g., centroid distance >0.3m), scale difference (e.g., area ratio <0.8 or >1.2), or morphological difference (e.g., contour similarity <0.7), then a geometric or positional deviation is determined. If a field-observed component belongs to the theoretically expected component set, but its visibility confidence is less than 0.3, and no candidate BIM component matching the field-observed component is found in the matching relationship, then an occlusion uncertainty difference is determined.

[0073] Furthermore, after determining the existence of occlusion-related uncertainties, the confidence level of missing differences is lowered based on their visibility confidence level. For example, a basic missing difference score for the component can be calculated first based on the overlap between the expected observation and the on-site observation (e.g., a basic score of 1 when the overlap is 0), and then this score is combined with the visibility confidence level. Specifically, when the visibility confidence level is low, the missing difference score for the component is reduced accordingly, and such components are marked as "occlusion-related uncertainties." In subsequent difference ranking, early warning prompts, and rectification task generation processes, occlusion-related uncertainties are given a lower alarm priority than those confirmed as missing differences to reduce false alarms caused by occlusion.

[0074] The above four types of differences comprehensively cover the core deviations in construction scenarios, and the judgment rules combine geometric features and temporal logic, which greatly improves accuracy.

[0075] In this embodiment, adjusting the detected differences using the 4D process timing constraint model as a constraint to obtain the target difference detection result includes: For each candidate BIM component identified as having a missing difference, the relationship between the process to which the candidate BIM component belongs and the completed processes is queried in the time sequence constraint diagram. If the current time has not reached the earliest start time of the process or there is no reachable path, the candidate BIM component is not identified as having a missing difference. If the candidate BIM component has a corresponding on-site observation component, it is identified as having a difference that occurred ahead of schedule or unexpectedly. Alternatively, if the process to which the candidate BIM component belongs is located on the core route or has multiple strongly dependent processes, and the current time exceeds the planned end time, the corresponding difference confidence is multiplied by a preset weight and marked as high priority. When it is determined from the timing constraint diagram that the process to which the component with geometric or positional deviation belongs is in a parallel relationship, the overlap threshold and the scale difference threshold are relaxed, and the corresponding component is re-detected for geometric or positional deviation based on the relaxed overlap threshold and the scale difference threshold. The differences after integration and adjustment are used to obtain the target difference detection results.

[0076] For example: when querying the relationship between the process to which a candidate BIM component belongs and the completed processes in the time-series constraint diagram, if the current time has not reached the earliest start time of the process or there is no reachable path (such as the previous process not being completed), it is not judged as a missing difference. It is only judged as an early or unplanned difference when there is on-site observation, thus realizing the elimination of time-series reachability; if the process to which the component belongs is parallel, the overlap threshold is relaxed to 0.2 and the scale difference threshold is relaxed to 0.7-1.3 to avoid local incompleteness or slight deviation being mistakenly judged as serious differences, thus adjusting the tolerance for parallel process matching; if the process to which the component belongs is located in the critical core path (such as the core process that affects multiple subsequent processes) or has multiple strong previous dependent processes, and the current time exceeds the planned end time, if it is judged as a missing difference, the difference confidence is multiplied by 1.5 (the weight can be customized) and marked as high priority.

[0077] With the assistance of the aforementioned 4D time-series constraint diagram, the false alarm rate (e.g., tolerance adjustment for parallel processes) and the false negative rate (e.g., weighting of processes with strong pre-dependent dependencies) are further reduced. Furthermore, the 4D process time-series constraint diagram plays a role in the difference determination process, including screening the comparison range, adaptively adjusting the matching threshold, and dynamically weighting the difference confidence level. It can simultaneously consider both the construction time-series logic and the geometric observation results of the images, thus improving the accuracy and engineering practicality of difference detection.

[0078] This embodiment closely links the difference determination results with the process and time, providing a clear basis for subsequent explanation and rectification.

[0079] In this embodiment, after obtaining the target difference detection results, the difference judgment results can be organized to generate a structured difference list. The structured difference list may include information such as difference type (e.g., missing difference), component ID, work area (e.g., first floor east unit) and floor (first floor), difference confidence level (e.g., 0.95), and corresponding process (e.g., first floor column pouring).

[0080] Of course, depending on the actual file format requirements, the target difference detection results can also be encapsulated into a data packet with semantic tags. This data packet may include fields such as: component ID, associated process code, difference type (e.g., missing, premature, deviation), difference score, and timestamp.

[0081] In this embodiment, image overlay technology can be used to highlight the projection area and difference area of ​​the BIM component (such as missing differences marked with a red box and geometric deviations marked with a yellow box) on the original construction site image to intuitively present the location and shape of the differences.

[0082] In the above embodiments, the structured discrepancy list and visualization results reduce the cost of manual review and make it easier for managers to quickly locate and understand the discrepancies.

[0083] In this embodiment, the component identifier, process, difference type, and difference confidence level, etc. (or the encapsulated data packet) in the structured difference list can also be written back to the 4D progress management database connected to the 4D planning system by calling the specified API (Application Programming Interface).

[0084] The 4D planning system is used to manage construction schedules and associate them with BIM 3D components. It has functions such as schedule preparation, process-BIM mapping, and 4D visualization.

[0085] For example, the 4D planning system can use this data to perform schedule deviation statistics (e.g., the "first-floor column pouring process" is 20% behind schedule due to the absence of 5 components), critical path analysis (e.g., the missing differences affect the subsequent "first-floor beam formwork" process), and automatically assign rectification tasks (e.g., notifying construction teams to supplement the construction of the corresponding components). For "early construction" issues determined to have discrepancies occurring ahead of schedule, the system automatically triggers an update to the "actual start time" field of the corresponding process node. For "high-priority missing" determinations, the system automatically calls the message notification service, pushing the original construction site image with highlighted coordinates and the BIM comparison map to the mobile terminal of the WBS-related responsible person, and generating a pending rectification work order. Through the above processing, management automation can be better achieved.

[0086] The 4D planning system can be a stand-alone 4D construction schedule planning software or a 4D schedule management subsystem integrated into a project management platform or a smart construction site platform.

[0087] The aforementioned feedback loop achieves deep integration of discrepancy detection results with construction progress management, transforming passive detection into proactive rectification, promoting dynamic optimization of the construction process, improving the automation level of project management, and reducing the workload of manual statistics and task assignment.

[0088] As can be seen from the above technical solutions, this invention can construct a 4D process sequence constraint model based on BIM model, construction schedule plan, and work breakdown structure process data, and for the first time transforms process sequence logic into a computable constraint model; based on pose information and visibility confidence, multiple BIM candidate components are selected from each BIM component in the theoretically expected component set, and a matching relationship is established between each on-site observed component and each BIM candidate component. Through multi-level filtering, the matching search space is significantly reduced, and the subsequent matching efficiency is significantly improved. Moreover, the matching relationship establishes a connection bridge between on-site observed components and BIM components, laying the foundation for difference determination; using the 4D process sequence constraint model as a constraint, difference detection is performed based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and multiple on-site observed components, effectively reducing the false alarm rate and false negative rate. Furthermore, the difference determination results are closely related to the process and time, providing a clear basis for subsequent explanation and rectification.

[0089] like Figure 2 The diagram shown is a functional block diagram of a preferred embodiment of the component difference detection device based on process constraints of the present invention. The component difference detection device 11 based on process constraints includes an acquisition unit 110, a construction unit 111, a projection unit 112, an identification unit 113, a screening unit 114, and a detection unit 115. The module / unit referred to in this invention refers to a series of computer program segments that can be executed by a processor and perform a fixed function, and are stored in memory. In this embodiment, the functions of each module / unit will be described in detail in subsequent embodiments.

[0090] The acquisition unit 110 is used to acquire the BIM model, construction schedule and work breakdown structure process data, and the current construction site image sequence in response to the component difference detection command. The construction unit 111 is used to construct a 4D process timing constraint model based on the BIM model, the construction schedule plan and the work breakdown structure process data, and to construct the theoretically expected component set and the theoretically unexpected component set at each time point based on the 4D process timing constraint model. The acquisition unit 110 is also used to acquire target scene images and acquire the pose information of the camera corresponding to the current moment image; The projection unit 112 is used to project each BIM component in the theoretically expected component set at the current moment onto the target site image according to the pose information to obtain the BIM component view projection, and generate the expected observation area and visibility confidence of each BIM component according to the BIM component view projection. The identification unit 113 is used to perform instance identification on the target scene image to obtain multiple scene observation components; The filtering unit 114 is used to filter multiple BIM candidate components from each BIM component in the theoretically expected component set according to the pose information and the visibility confidence, and to establish a matching relationship between each on-site observed component and each BIM candidate component. The detection unit 115 is used to perform difference detection based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observation components, using the 4D process sequence constraint model as a constraint, to obtain the target difference detection result.

[0091] As can be seen from the above technical solutions, this invention can construct a 4D process sequence constraint model based on BIM model, construction schedule plan, and work breakdown structure process data, and for the first time transforms process sequence logic into a computable constraint model; based on pose information and visibility confidence, multiple BIM candidate components are selected from each BIM component in the theoretically expected component set, and a matching relationship is established between each on-site observed component and each BIM candidate component. Through multi-level filtering, the matching search space is significantly reduced, and the subsequent matching efficiency is significantly improved. Moreover, the matching relationship establishes a connection bridge between on-site observed components and BIM components, laying the foundation for difference determination; using the 4D process sequence constraint model as a constraint, difference detection is performed based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and multiple on-site observed components, effectively reducing the false alarm rate and false negative rate. Furthermore, the difference determination results are closely related to the process and time, providing a clear basis for subsequent explanation and rectification.

[0092] like Figure 3 The diagram shown is a schematic representation of the structure of a computer device that implements a component difference detection method based on process constraints according to the present invention.

[0093] The computer device 1 may include a memory 12, a processor 13, and a bus (the arrow in the figure represents the bus), and may also include a computer program stored in the memory 12 and executable on the processor 13, such as a component difference detection program based on process constraints.

[0094] Those skilled in the art will understand that the schematic diagram is merely an example of computer device 1 and does not constitute a limitation on computer device 1. Computer device 1 can be either a bus topology or a star topology. Computer device 1 may also include more or fewer other hardware or software than shown in the diagram, or different component arrangements. For example, computer device 1 may also include input / output devices, network access devices, etc.

[0095] It should be noted that the computer device 1 described is merely an example. Other existing or future electronic products that are adaptable to this invention should also be included within the scope of protection of this invention and are incorporated herein by reference.

[0096] The memory 12 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 12 can be an internal storage unit of the computer device 1, such as a portable hard drive of the computer device 1. In other embodiments, the memory 12 can be an external storage device of the computer device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the computer device 1. Furthermore, the memory 12 can include both internal storage units and external storage devices of the computer device 1. The memory 12 can be used not only to store application software and various types of data installed on the computer device 1, such as the code of a component difference detection program based on process constraints, but also to temporarily store data that has been output or will be output.

[0097] In some embodiments, the processor 13 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits packaged with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 13 is the control unit of the computer device 1, connecting various components of the computer device 1 via various interfaces and lines. It executes programs or modules stored in the memory 12 (e.g., executing a component difference detection program based on process constraints) and calls data stored in the memory 12 to perform various functions of the computer device 1 and process data.

[0098] The processor 13 executes the operating system of the computer device 1 and various installed applications. The processor 13 executes the applications to implement the steps in the above embodiments of the component difference detection method based on process constraints, for example... Figure 1 The steps are shown.

[0099] For example, the computer program may be divided into one or more modules / units, which are stored in the memory 12 and executed by the processor 13 to complete the present invention. The one or more modules / units may be a series of computer-readable instruction segments capable of performing specific functions, which describe the execution process of the computer program in the computer device 1. For example, the computer program may be divided into an acquisition unit 110, a construction unit 111, a projection unit 112, an identification unit 113, a filtering unit 114, and a detection unit 115.

[0100] The integrated unit implemented as a software functional module described above can be stored in a computer-readable storage medium. This software functional module, stored in a storage medium, includes several instructions to cause a computer device (which may be a personal computer, computer equipment, or network device, etc.) or processor to execute portions of the component difference detection method based on process constraints described in the various embodiments of the present invention.

[0101] If the modules / units integrated in the computer device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware devices. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above.

[0102] The computer program includes computer program code, which may be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory, etc.

[0103] Furthermore, the computer-readable storage medium may primarily include a stored program area and a stored data area, wherein the stored program area may store the operating system, an application program required for at least one function, etc.; and the stored data area may store data created based on the use of blockchain nodes, etc.

[0104] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0105] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, in... Figure 3 The bus is represented by only one straight line, but this does not mean that there is only one bus or one type of bus. The bus is configured to enable communication between the memory 12 and at least one processor 13, etc.

[0106] Although not shown, the computer device 1 may also include a power supply (such as a battery) to power various components. Preferably, the power supply can be logically connected to the at least one processor 13 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The computer device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0107] Furthermore, the computer device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the computer device 1 and other computer devices.

[0108] Optionally, the computer device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the computer device 1 and to display a visual user interface.

[0109] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0110] It will be understood by those skilled in the art that Figure 3 The structure shown does not constitute a limitation on the computer device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0111] Combination Figure 1 The memory 12 in the computer device 1 stores multiple instructions to implement a component difference detection method based on process constraints, and the processor 13 can execute the multiple instructions to achieve the following: In response to component difference detection commands, acquire BIM model, construction schedule and work breakdown structure process data, and current construction site image sequence; A 4D process timing constraint model is constructed based on the BIM model, the construction schedule, and the work breakdown structure process data. The theoretical set of components that should appear and the theoretical set of components that should not appear at each time point are then constructed based on the 4D process timing constraint model. Acquire target scene images and acquire the pose information of the camera corresponding to the current moment image; Based on the pose information, each BIM component in the theoretically expected component set at the current moment is projected onto the target site image to obtain the BIM component view projection, and the expected observation area and visibility confidence of each BIM component are generated based on the BIM component view projection. Multiple on-site observation components were obtained by performing instance recognition on the target site image; Based on the pose information and the visibility confidence, multiple BIM candidate components are selected from each BIM component in the theoretically expected component set, and a matching relationship is established between each on-site observed component and each BIM candidate component. Using the 4D process sequence constraint model as a constraint, difference detection is performed based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observation components to obtain the target difference detection result.

[0112] Specifically, the processor 13's implementation method for the above instructions can be found in [reference needed]. Figure 1 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0113] It should be noted that all the data involved in this case was legally obtained.

[0114] If any AI models, software tools, or components not belonging to this company appear in the embodiments of this invention, they are merely illustrative examples and do not represent actual use. All user personal information involved in the embodiments of this invention has been obtained by an entity authorized (with the knowledge and consent) or fully authorized by all parties through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with relevant laws and regulations and do not violate public order and good morals.

[0115] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0116] This invention can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0117] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0119] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0120] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0121] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices described in this invention can also be implemented by a single unit or device through software or hardware. Terms such as "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A component difference detection method based on process constraints, characterized in that, The component difference detection method based on process constraints includes: In response to component difference detection commands, acquire BIM model, construction schedule and work breakdown structure process data, and current construction site image sequence; A 4D process timing constraint model is constructed based on the BIM model, the construction schedule, and the work breakdown structure process data. The theoretical set of components that should appear and the theoretical set of components that should not appear at each time point are then constructed based on the 4D process timing constraint model. Acquire target scene images and acquire the pose information of the camera corresponding to the current moment image; Based on the pose information, each BIM component in the theoretically expected component set at the current moment is projected onto the target site image to obtain the BIM component view projection, and the expected observation area and visibility confidence of each BIM component are generated based on the BIM component view projection. Multiple on-site observation components were obtained by performing instance recognition on the target site image; Based on the pose information and the visibility confidence, multiple BIM candidate components are selected from each BIM component in the theoretically expected component set, and a matching relationship is established between each on-site observed component and each BIM candidate component. Using the 4D process sequence constraint model as a constraint, difference detection is performed based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observation components to obtain the target difference detection result.

2. The component difference detection method based on process constraints as described in claim 1, characterized in that, The step of constructing a 4D process sequence constraint model based on the BIM model, the construction schedule, and the work breakdown structure process data includes: Extract each process and the relationships between every two processes from the construction schedule and the work breakdown structure process data; wherein, the relationships between every two processes include preceding relationships, subsequent relationships, parallel relationships, and conflict relationships; Using each process as a node, and constructing edges based on the relationship between every two processes, a timing constraint graph is obtained; The planned time interval for each process is determined based on the construction schedule and the work breakdown structure process data. Each BIM component is extracted from the BIM model, and its status is mapped according to the planned time interval of each process; wherein, the process status of each BIM component includes not under construction, under construction, and under construction. The state of each BIM component is added to the timing constraint diagram as an attribute association object of each node to obtain the 4D process timing constraint model.

3. The component difference detection method based on process constraints as described in claim 2, characterized in that, The theoretically expected and theoretically unappearable component sets for each moment, constructed based on the 4D process timing constraint model, include: For each BIM component at any given time, the planned time interval of the corresponding process of the BIM component is obtained according to the 4D process sequence constraint model, and the buffer window is obtained. Add the buffer window to the planned time interval to obtain a buffer space, and check whether the moment is within the buffer space; When the moment is detected to be within the buffer zone, the timing constraint diagram is used to detect whether the preceding process of the process corresponding to the BIM component has been completed or is in a parallel state. When it is detected that the preceding process of the process corresponding to the BIM component has been completed or is in a parallel state, the BIM component is added to the theoretically expected component set; or When it is detected that the preceding process of the process corresponding to the BIM component is not completed and is not in a parallel state, the BIM component is added to the theoretically non-existent component set.

4. The component difference detection method based on process constraints as described in claim 2, characterized in that, The step of generating the expected observation area and visibility confidence of each BIM component based on the view projection of the BIM component includes: Based on the view projection of the BIM component, generate the contour mask, bounding box and visible face of each BIM component, and generate the desired observation area of ​​each BIM component based on the contour mask, bounding box and visible face of each BIM component. For each BIM component, extract the exposed surface of the BIM component facing the current camera, and extract a preset number of sampling points from the exposed surface; Starting from the center of the camera, ray detection is performed along the line of sight at each sampling point to determine whether each ray intersects with other components before reaching the corresponding sampling point; The sampling points corresponding to rays that intersect with other components are determined as geometric occlusion sampling points, and the sampling points corresponding to rays that do not intersect with other components are determined as geometrically visible sampling points. The geometric visibility rate of the BIM component is obtained by calculating the quotient of the number of geometrically visible sampling points and the preset number. Identify each occlusion component involved in the occlusion, and determine the process state of each occlusion component according to the 4D process timing constraint model; The occlusion rule library is retrieved, and the visibility adjustment factor for each occlusion component is obtained by traversing the occlusion rule library based on the process status of each occlusion component; wherein, the occlusion rule library is used to store the correspondence between component type, process status, and visibility adjustment factor; The minimum value of the visibility adjustment factor of all occluded components is obtained as the target visibility adjustment factor of the BIM component; The visibility confidence level of the BIM component is obtained by multiplying the geometric visibility rate of the BIM component with the target visibility adjustment factor.

5. The component difference detection method based on process constraints as described in claim 2, characterized in that, The step of selecting multiple BIM candidate components from each BIM component in the theoretically expected component set based on the pose information and the visibility confidence level includes: Based on the camera extrinsic parameters and view frustum parameters in the pose information, each BIM component is projected to the current viewpoint to obtain the projection result. Based on the projection result, components that do not intersect with the view frustum or whose projected area is smaller than a preset area are removed from each BIM component to obtain a subset of components that should appear in the field of view. BIM components with visibility confidence scores lower than the confidence threshold are removed from the subset of components that should appear within the field of view to obtain the plurality of BIM candidate components.

6. The component difference detection method based on process constraints as described in claim 5, characterized in that, The step of using the 4D process sequence constraint model as a constraint, and performing difference detection based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observed components, yields the following target difference detection results: For each candidate BIM component, if the overlap between the candidate BIM component and all field observation components is less than the overlap threshold or no field observation component matching the candidate BIM component is found in the matching relationship, it is determined that there is a missing difference. For each field observation component, if the field observation component belongs to the set of components that should not appear according to theory, and the identification confidence of the field observation component is greater than a preset threshold, it is determined that there is a difference between early appearance and unplanned appearance. According to the matching relationship, when it is detected that the overlap between a candidate BIM component and the matching field observation component is greater than or equal to the overlap threshold, and the displacement is greater than a preset displacement, the scale difference is greater than the scale difference threshold, and the morphological difference is greater than a preset amount, it is determined that there is a geometric or positional deviation. For each field observation component, if the field observation component belongs to the theoretically expected component set, the visibility confidence of the field observation component is less than the confidence threshold, and no candidate BIM component matching the field observation component is found in the matching relationship, it is determined that there is an occlusion uncertainty difference, and the alarm priority of the field observation component is reduced. The detected differences are adjusted using the 4D process timing constraint model as a constraint to obtain the target difference detection result.

7. The component difference detection method based on process constraints as described in claim 6, characterized in that, The step of adjusting the detected differences using the 4D process timing constraint model as a constraint to obtain the target difference detection result includes: For each candidate BIM component identified as having a missing difference, the relationship between the process to which the candidate BIM component belongs and the completed processes is queried in the time sequence constraint diagram. If the current time has not reached the earliest start time of the process or there is no reachable path, the candidate BIM component is not identified as having a missing difference. If the candidate BIM component has a corresponding on-site observation component, it is identified as having a difference that occurred ahead of schedule or unexpectedly. Alternatively, if the process to which the candidate BIM component belongs is located on the core route or has multiple strongly dependent processes, and the current time exceeds the planned end time, the corresponding difference confidence is multiplied by a preset weight and marked as high priority. When it is determined from the timing constraint diagram that the process to which the component with geometric or positional deviation belongs is in a parallel relationship, the overlap threshold and the scale difference threshold are relaxed, and the corresponding component is re-detected for geometric or positional deviation based on the relaxed overlap threshold and the scale difference threshold. The differences after integration and adjustment are used to obtain the target difference detection results.

8. A component difference detection device based on process constraints, characterized in that, The component difference detection device based on process constraints includes: The acquisition unit is used to acquire the BIM model, construction schedule and work breakdown structure process data, and the current construction site image sequence in response to the component difference detection command. The construction unit is used to construct a 4D process timing constraint model based on the BIM model, the construction schedule plan and the work breakdown structure process data, and to construct the theoretically expected component set and the theoretically unexpected component set at each time point based on the 4D process timing constraint model. The acquisition unit is also used to acquire target scene images and acquire the pose information of the camera corresponding to the current moment image; The projection unit is used to project each BIM component in the theoretically expected component set at the current moment onto the target site image according to the pose information to obtain the BIM component view projection, and generate the expected observation area and visibility confidence of each BIM component according to the BIM component view projection. The identification unit is used to perform instance identification on the target scene image to obtain multiple scene observation components; The filtering unit is used to filter multiple BIM candidate components from each BIM component in the theoretically expected component set based on the pose information and the visibility confidence, and to establish a matching relationship between each on-site observed component and each BIM candidate component. The detection unit is used to perform difference detection based on the expected observation area and visibility confidence of each candidate BIM component, the matching relationship, and the multiple on-site observation components, using the 4D process sequence constraint model as a constraint, to obtain the target difference detection result.

9. A computer device, characterized in that, The computer device includes: A memory that stores at least one instruction; and a processor that executes the instructions stored in the memory to implement the component difference detection method based on process constraints as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, which is executed by a processor in a computer device to implement the component difference detection method based on process constraints as described in any one of claims 1 to 7.

Images