BIM model standard layer identification method and system based on plane projection comparison, program product and storage medium

Abstract

The invention provides a BIM model standard layer identification method and system based on plane projection comparison, a program product and a storage medium, and relates to the technical field of BIM model identification. Cutting the BIM model according to floor boundaries to obtain a plurality of floor models, then dividing the floor models with the same floor height into one group, performing vertical projection under the same coordinate system for each floor model group, converting a three-dimensional floor model into a two-dimensional floor projection image, and performing three-dimensional projection on the two-dimensional floor projection image; the data form is simplified while preserving information based on subsequent pixel comparison. Cutting the image according to a preset grid to generate a plurality of image units, comparing the image units at the same grid position in the same floor model group in pairs to judge a standard floor, and efficiently and accurately completing standard floor identification. The problems of misjudgment and workload increase in application caused by lack of standard layer information when the BIM model is converted or imported in the past are solved, and the overall working efficiency and the standard layer identification accuracy are improved.

Description

Technical Field

[0001] This application relates to the field of BIM model recognition technology, and in particular to a method, system, program product and storage medium for BIM model standard layer recognition based on planar projection comparison. Background Technology

[0002] Currently, when converting BIM models to IFC format or importing them directly into other platforms, standard layer attribute information is not retained for model components. This leads to a lack of data in the subsequent application of the BIM model, potentially causing misjudgments and increasing workload. Summary of the Invention

[0003] This application provides a method, system, program product, and storage medium for identifying standard layers in BIM models based on planar projection comparison. This method addresses the problems of judgment errors and increased workload caused by the lack of standard layer information when converting or importing BIM models, thereby improving overall work efficiency and the accuracy of standard layer identification.

[0004] Firstly, this application provides a method for identifying standard floors in BIM models based on planar projection comparison, comprising: acquiring the BIM model to be analyzed; cutting the BIM model according to floor boundaries to obtain multiple floor models, each floor model being a three-dimensional sub-model corresponding to a single building floor; grouping floor models with the same floor height into several floor model groups; for each floor model group, vertically projecting each floor model within the group onto the same coordinate system to generate a corresponding two-dimensional floor projection image for each floor model; cutting the floor projection image into several image units according to a preset grid; within the same floor model group, comparing the image units corresponding to different floor models and located at the same grid position pairwise; if all corresponding image units in the floor model are identical, then the floor model is determined to be a standard floor.

[0005] By employing the above technical solution, the BIM model to be analyzed is obtained. Then, the BIM model is cut according to floor boundaries to obtain multiple floor models. Floor models with the same floor height are grouped together, and preliminary classification based on floor height reduces workload and prepares for further comparison within the same group. For each floor model group, vertical projection is performed in the same coordinate system, converting the 3D floor model into a 2D floor projection image, simplifying the data format while preserving information based on subsequent pixel comparisons. Then, the image is cut according to a preset grid to generate several image units. Within the same floor model group, image units at the same grid position are compared pairwise to determine the standard floor. This efficiently and accurately completes standard floor identification, solving the problems of judgment errors and increased workload caused by missing standard floor information during previous BIM model conversions or imports, thus improving overall work efficiency and the accuracy of standard floor identification.

[0006] In conjunction with some embodiments of the first aspect, in some embodiments, after the step of determining that the floor model is a standard floor if all corresponding image units in the floor model are the same, the method further includes: if there is at least one pair of different image units, the different image units are recorded as difference image units; for each difference image unit, backtracking to the corresponding original three-dimensional space range and obtaining the original three-dimensional space components within the space range; performing three-dimensional geometric similarity analysis on the three-dimensional space components corresponding to the difference image units; when the three-dimensional space components are geometrically the same or proportionally related, it is determined to be a standard floor with a variable cross-section; generating and outputting the standard floor recognition result containing standard floor information.

[0007] By adopting the above technical solution, when determining whether a floor model is a standard floor, if at least one pair of image units are different, the system then backtracks to the corresponding original 3D spatial range and obtains the original 3D spatial components within that range, allowing the analysis to move beyond surface image differences. Next, a 3D geometric similarity analysis is performed on the 3D spatial components corresponding to the differing image units. In practical use, some floor model groups may appear different after vertical projection, but they are actually special standard floors. Through geometric similarity analysis of the components, when the 3D spatial components are geometrically identical or proportionally related, they are identified as standard floors with variable cross-sections. This overcomes the limitations of relying solely on image comparison. Even if differences occur during the image comparison stage, subsequent 3D component analysis can uncover those special standard floor cases, avoiding the problem of misclassifying special standard floors as non-standard floors and improving the reliability of standard floor application throughout the entire cycle.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, when three-dimensional spatial components are geometrically identical or proportionally related, the step of determining a standard layer with variable cross-section specifically includes: when the geometric topology of the three-dimensional spatial components is the same and the dimensional parameters have a constant scaling ratio in the horizontal or vertical direction, then it is determined to be a standard layer with variable cross-section.

[0009] By adopting the above technical solution, the rule that three-dimensional spatial components are judged as standard layers with variable cross-sections when they are geometrically identical or proportionally related is clarified. Furthermore, it is further refined to the point that when the geometric topology of three-dimensional spatial components is the same and the dimensional parameters have a constant scaling ratio in the horizontal or vertical direction, they are judged as standard layers with variable cross-sections. This is more in line with the actual situation and provides a clear and quantifiable standard for judgment.

[0010] In conjunction with some embodiments of the first aspect, before the step of grouping floor models with the same floor height into several floor model groups, the method further includes: sequentially pairing all floor models into corresponding floor groups; comparing the floor heights of the floor groups; if the floor height of the current floor model is different from the floor heights of all other floor models, then the current floor model is determined to be a non-standard floor, the current floor model is any floor model, and the other floor models are floor models excluding the current floor model.

[0011] By adopting the above technical solution, before grouping floor models with the same floor height into groups, all floor models are first paired sequentially to form corresponding floor pairs. This allows for an initial screening of all floors, quickly comparing the floor heights of adjacent floors or two floors in sequence. Subsequent floor height comparisons directly identify floors with different floor heights from other floors, classifying them as non-standard floors, thus achieving preliminary screening of non-standard floors. This also provides a basis for accurately grouping floor models with the same floor height in the subsequent process, improving the overall efficiency of distinguishing between standard and non-standard floors.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, the step of grouping floor models with the same floor height into several floor model groups specifically includes: during the floor height comparison process, collecting and summarizing other floor models with the same floor height as the current floor model; and combining the collected and summarized other floor models with the current floor model to form the floor model group corresponding to the current floor model.

[0013] By adopting the above technical solution, during the floor height comparison process, other floor models with the same floor height as the current floor model are collected and summarized. These collected and summarized floor models are then combined with the current floor model to form the floor model group corresponding to the current floor model. This allows for the rapid and accurate grouping of similar floors based on the key attribute of floor height, forming sets and improving the accuracy of the entire process.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, before the step of cutting the BIM model according to the floor boundaries to obtain multiple floor models, wherein the floor model is a three-dimensional sub-model corresponding to a single building floor, the method further includes: converting the BIM model into a three-dimensional wireframe model.

[0015] By adopting the above technical solution, before cutting the BIM model according to floor boundaries, the BIM model is first converted into a 3D wireframe model. In the entire standard floor identification process, it is not necessary to use all the information contained in the complete and complex BIM model; the 3D wireframe model alone is sufficient for subsequent operations. The 3D wireframe model outlines the basic structural framework of the building with simple lines, clearly presenting the boundary contours between floors and the approximate shape of each major component. Subsequent operations, such as grouping floor models with the same floor height, performing vertical projection, and comparing image units, can all be smoothly carried out based on the relatively simple floor model derived from the 3D wireframe model. This avoids unnecessary analytical burdens due to redundant model information and does not affect the accuracy of the final standard floor identification.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, after determining a standard layer with variable cross-section when the three-dimensional spatial components are geometrically identical or proportionally related, the method further includes: determining a non-standard layer when the three-dimensional spatial components are geometrically dissimilar or not proportionally related.

[0017] By adopting the above technical solution, when the geometric proportions of three-dimensional spatial components are not the same or do not conform to the proportions, they are determined to be non-standard floors. This rule can accurately screen out the floors where components that do not meet the characteristics of standard floors are located.

[0018] Secondly, this application provides a BIM model standard layer identification system based on planar projection comparison. The BIM model standard layer identification system based on planar projection comparison includes: one or more processors and a memory; the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and one or more processors call the computer instructions to cause the BIM model standard layer identification system based on planar projection comparison to perform the method described in the first aspect and any possible implementation of the first aspect.

[0019] Thirdly, this application provides a computer program product containing instructions that, when the computer program product is run on a BIM model standard layer identification system based on planar projection comparison, causes the BIM model standard layer identification system based on planar projection comparison to perform the method described in the first aspect and any possible implementation thereof.

[0020] Fourthly, this application provides a computer-readable storage medium including instructions that, when executed on a BIM model standard layer identification system based on planar projection comparison, cause the BIM model standard layer identification system based on planar projection comparison to perform the method described in the first aspect and any possible implementation thereof.

[0021] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. Obtain the BIM model to be analyzed, then cut the BIM model according to floor boundaries to obtain multiple floor models. Group floor models with the same floor height together, using floor height as a preliminary classification feature to reduce workload and prepare for further comparisons within the same group. For each floor model group, perform vertical projection in the same coordinate system to convert the 3D floor model into a 2D floor projection image, simplifying the data format while preserving information based on subsequent pixel comparisons. Then, cut the image according to a preset grid to generate several image units. Within the same floor model group, compare image units at the same grid position pairwise to determine the standard floor. This efficiently and accurately completes standard floor identification, solving the problems of judgment errors and increased workload caused by missing standard floor information during previous BIM model conversions or imports, thus improving overall work efficiency and the accuracy of standard floor identification.

[0022] 2. When determining whether a floor model is a standard floor, if at least one pair of image units are different, the system then backtracks to the corresponding original 3D spatial range and obtains the original 3D spatial components within that range, allowing the analysis to move beyond surface image differences. Next, a 3D geometric similarity analysis is performed on the 3D spatial components corresponding to the differing image units. In practical use, some floor model groups may appear different after vertical projection, but they are actually special standard floors. Through geometric similarity analysis of the components, when the 3D spatial components are geometrically identical or proportionally related, they are identified as standard floors with variable cross-sections. This overcomes the limitations of relying solely on image comparison. Even if differences occur during the image comparison stage, subsequent 3D component analysis can uncover those special standard floor cases, avoiding the problem of misclassifying special standard floors as non-standard floors and improving the reliability of standard floor application throughout the entire cycle. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of an exemplary application scenario of the BIM model standard layer identification method based on planar projection comparison in the embodiments of this application; Figure 2 This is a flowchart illustrating a BIM model standard layer identification method based on planar projection comparison in an embodiment of this application. Figure 3 This is another flowchart illustrating the BIM model standard layer identification method based on planar projection comparison in the embodiments of this application; Figure 4 This is an exemplary hardware structure diagram of a BIM model standard layer identification system based on planar projection comparison in the embodiments of this application. Detailed Implementation

[0024] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0025] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0026] Please see Figure 1 and 2 , Figure 1 This is a schematic diagram illustrating an exemplary application scenario of the BIM model standard layer identification method based on planar projection comparison in the embodiments of this application. Figure 2 This is a flowchart illustrating a BIM model standard layer identification method based on planar projection comparison in an embodiment of this application. A method for identifying standard layers in a BIM model based on planar projection comparison, comprising: S201. Obtain the BIM model to be analyzed, and cut the BIM model according to the floor boundaries to obtain multiple floor models. Each floor model is a three-dimensional sub-model corresponding to a single building floor. Before starting standard floor identification, it is necessary to separate the models of each floor from the complete BIM model for subsequent analysis.

[0027] In this context, the floor boundary refers to the spatial boundary of each floor in a building. It should be noted that this includes the horizontal boundary, which is the boundary that divides the different floor areas in the horizontal direction. It is mainly defined by the horizontal structural components in the building, such as floor slabs and mezzanine slabs, which enclose a unique spatial area for each floor on a horizontal level.

[0028] In some embodiments, floor boundaries can be identified based on features. For example, floor slabs are key features. In a BIM model, floor slabs have specific elevation attributes and planar geometry. By reading the bottom and top elevation values ​​of each floor slab, its vertical position range can be determined. The areas covered by adjacent floor slabs on the horizontal plane constitute the lateral boundary range of the corresponding floor. Based on these features, the lateral boundary of the floor can be accurately defined under different conditions.

[0029] In other embodiments, floor boundaries can be determined based on conventional features of building design. For example, vertical circulation spaces such as staircases and elevators in a building typically run through multiple floors, and the midline between the openings on each floor is the floor boundary.

[0030] Specifically, the model cutting function is used to divide the BIM model along the floor boundaries, making each floor an independent three-dimensional sub-model, which facilitates subsequent operation and analysis of individual floors.

[0031] It should be noted that, considering that subsequent operations will focus on key information such as the building's structural framework and spatial layout, and do not require all the complex details contained in the BIM model, it will be converted into a simpler 3D wireframe model. This will simplify the subsequent analysis process, reduce unnecessary information interference, and improve the efficiency of the overall standard floor identification work.

[0032] Therefore, in some embodiments, step S201 is followed by: converting the BIM model into a three-dimensional wireframe model.

[0033] Among them, a 3D wireframe model refers to a frame structure of a 3D object described by lines. It uses only a series of line segments to outline the shape of the object and show its basic spatial form and the topological relationship between its parts.

[0034] It is easy to imagine that the models used in the subsequent steps are all 3D wireframe models.

[0035] As can be seen, before cutting the BIM model according to floor boundaries, it is necessary to first convert the BIM model into a 3D wireframe model. In the entire standard floor identification process, it is not necessary to use all the information contained in the complete and complex BIM model; the 3D wireframe model alone can meet the needs of subsequent operations. The 3D wireframe model outlines the basic structural framework of the building with simple lines. It clearly presents the boundary contours between floors and the approximate shape of each major component. Subsequent operations, such as grouping floor models with the same floor height, performing vertical projection, and comparing image units, can all be carried out smoothly based on the relatively simple floor model derived from the 3D wireframe model. This avoids unnecessary analysis burden due to redundant model information and does not affect the accuracy of the final standard floor identification.

[0036] Please see Figure 1 The left side of the image shows a schematic diagram of the floors. First, the overall building model is cut according to the floor boundaries, making each floor an independent entity. On this basis, these independent floor models are converted into three-dimensional wireframe models. The value in the middle is the floor number of the floor model.

[0037] S202. Divide the floor models with the same floor height into a group to obtain several floor model groups; Here, floor height represents the vertical distance between different floors of a building. A floor model group refers to a collection of floor models with the same floor height.

[0038] Specifically, all floor models are traversed to obtain the floor height data for each model. Floor models with the same floor height are grouped together to form a floor model group. After processing all floor models in this way, several floor model groups are obtained. In this way, the floor models are initially filtered and grouped according to the feature of floor height, reducing the scope and workload of subsequent comparisons.

[0039] In some embodiments, before step S202, the method further includes: a) sequentially pairing all floor models into corresponding floor groups; Specifically, starting with the first floor model, each model is selected sequentially and paired with the next floor model to form a floor pair. For example, the first floor model and the second floor model form a pair, then the second floor model and the third floor model form a pair, and so on, until all floor models are paired in sequence. In this way, all floor models are organized in pairs, facilitating subsequent comparisons of floor heights for each pair.

[0040] Please see Figure 1Here, the paired floors are marked with "number-number" to clearly show the combination relationship between floors. For example, "1-2" represents the pairing of the 1st floor and the 2nd floor. Then, in sequence, there is "1-3", which means that the 1st floor and the 3rd floor are paired. Following this pattern, a series of pairs will appear, such as "1-n" ("n" refers to different floor numbers, used to show various combinations with the 1st floor), and then other different floor combinations such as "2-3" will also be presented.

[0041] b) Compare the floor heights of the floor groups; c) If the floor height of the current floor model is different from the floor height of all other floor models, then the current floor model is determined to be a non-standard floor, the current floor model is any floor model, and the other floor models are floor models excluding the current floor model.

[0042] Specifically, each floor model is iterated through, and for the currently being processed floor model (i.e., the current floor model), its floor height is compared with the floor heights of all other floor models one by one. It is checked whether its floor height is different from the floor heights of every other floor model. If so, then the floor corresponding to the current floor model is determined to be a non-standard floor.

[0043] It should be noted that in this step, after determining that a certain floor model belongs to a non-standard floor, due to the logical setting of the entire BIM model standard floor identification method based on planar projection comparison, the floor model will no longer participate in subsequent steps aimed at identifying standard floors, such as vertical projection and image unit comparison.

[0044] It should be noted that as the subsequent steps are carried out, new non-standard layers may arise. Of course, there are additional steps and procedures to handle these non-standard layers, which will not be elaborated here.

[0045] As can be seen, before grouping floor models with the same floor height into a single group, all floor models are first paired sequentially to perform an initial screening of all floors, quickly comparing the floor heights of adjacent floors or two floors in sequence. Subsequent floor height comparisons directly identify floors with different floor heights from other floors, classifying them as non-standard floors, thus achieving preliminary screening of non-standard floors. This also provides a basis for accurately grouping floor models with the same floor height in the subsequent process, improving the overall efficiency of distinguishing between standard and non-standard floors.

[0046] In the above embodiment, a specific embodiment of step S202 is: d) During the floor height comparison process, collect and summarize other floor models with the same floor height as the current floor model; Specifically, when comparing floor heights in a floor group, once a current floor model is determined, the floor heights of other floor models are checked. By comparing previously recorded floor height data or retrieving floor height information, other floor models with the same floor height as the current floor model are selected. These selected floor models are then aggregated and collected in some way (e.g., added to the same list, set, or other data structure) to form a temporary set, which is convenient for later use to combine with the current floor model to form the corresponding floor model group.

[0047] e) Combine the collected and summarized other floor models with the current floor model to form the floor model group corresponding to the current floor model.

[0048] Specifically, the other floor models with the same floor height as the current floor model, collected and summarized in step d, are combined with the current floor model and used in a specific way (such as explicitly associating them in the data structure or setting them as a group in the software) to form a complete floor model group. This creates a set of floor models based on the characteristic of having the same floor height.

[0049] As can be seen, during the floor height comparison process, other floor models with the same floor height as the current floor model are collected and summarized, and these collected and summarized floor models are combined with the current floor model to form the floor model group corresponding to the current floor model. This allows for the rapid and accurate grouping of similar floors based on the key attribute of floor height, forming sets and improving the accuracy of the entire process.

[0050] S203. For each floor model group, project each floor model in the group vertically in the same coordinate system to generate a corresponding two-dimensional floor projection image for each floor model. The use of the same coordinate system ensures the uniformity and comparability of projections from different floor models. Vertical projection refers to the operation of projecting a floor model in three-dimensional space onto a two-dimensional plane along a vertical direction. A two-dimensional floor projection image is a planar image obtained after vertical projection of the floor model, containing the outline and layout information of the floor model in the vertical direction.

[0051] Specifically, for each group of floor models, the floor models within the group are selected sequentially. The selected floor models are placed in a pre-defined coordinate system, and then a projection algorithm is used to project the 3D components and geometric information of the floor models onto a specified 2D plane along the vertical direction, generating a corresponding 2D floor projection image. In this way, each floor model has a 2D image that reflects its vertical characteristics.

[0052] S204. Cut the floor projection image into several image units according to a preset grid; The preset grid refers to a pre-defined array of squares used to cut the image, based on actual needs and image resolution. An image unit is a small image patch obtained after the floor projection image is cut by the grid; each image unit contains partial information from the floor projection image.

[0053] Specifically, a grid is overlaid on the floor projection image, and the image is divided into several image units of the same or similar size according to the grid lines. In this way, the entire floor projection image is decomposed into multiple small blocks, which facilitates subsequent individual comparative analysis of each block.

[0054] It should be noted that the images can be cut into any shape, as long as the cutting method is the same for the floor projection images that need to be compared.

[0055] S205. Within the same floor model group, perform pairwise comparisons of image units corresponding to different floor models that are located at the same grid position. Here, "same floor model group" refers to a collection of floor models with the same floor height. "Same grid position" refers to the corresponding grid area in the projected images of different floor models, divided by a preset grid.

[0056] Specifically, for any two floor models within the same floor model group, their corresponding two-dimensional floor projection images are found. Based on a preset grid, image units at the same grid positions are sequentially extracted, and image comparison algorithms (such as calculating pixel differences and feature similarity) are used to compare these image units, recording the comparison results. By comparing all image units at the same grid positions pairwise, a comprehensive comparison of the similarity between floor models can be achieved.

[0057] In some embodiments, the right side of the figure shows the comparison. Suppose we have floor model 1 and floor model n, which together form a floor model group. The next step is to process them according to the standard procedure. First, the two floor models are vertically projected in the same coordinate system, converting the 3D floor models into 2D projected images, allowing their structure and features to be presented more intuitively. Next, one image unit is extracted from each of the two floor model's corresponding projected images, and these two extracted image units are compared. This comparative analysis helps to understand the similarity between the two floor models in corresponding parts, providing a reference for subsequent work such as determining whether the floor model group meets specific requirements.

[0058] S206. If all corresponding image units in the floor model are identical, then the floor model is determined to be a standard floor.

[0059] Specifically, iterate through all the results of comparing a floor model within the same floor model group with other floor models, and check whether all corresponding image units are the same. If all corresponding image units are the same, it means that the floor model is consistent with other floor models in the group in terms of structure and layout, and can be determined as a standard floor; if there is at least one different corresponding image unit, it cannot be determined as a standard floor.

[0060] As can be seen, obtaining the BIM model to be analyzed, then cutting the BIM model according to floor boundaries to obtain multiple floor models, and then grouping floor models with the same floor height together for preliminary classification based on floor height, can reduce workload and prepare for further comparison within the same group. For each floor model group, vertical projection is performed in the same coordinate system, transforming the 3D floor model into a 2D floor projection image, simplifying the data format while preserving information based on subsequent pixel comparison. Then, the image is cut according to a preset grid to generate several image units. Within the same floor model group, image units at the same grid position are compared pairwise to determine the standard floor. This efficiently and accurately completes the standard floor identification, solving the problems of judgment errors and increased workload caused by the lack of standard floor information when converting or importing BIM models in the past, and improving the overall work efficiency and the accuracy of standard floor identification.

[0061] In practical use, some floor model groups belong to a special category of standard floors. However, in the standard floor identification process, after the vertical projection step, when comparing the image units generated by the projection pairwise, different judgments are often made, making it impossible to accurately identify these special standard floors. This reduces the accuracy of standard floor identification to some extent.

[0062] Please see Figure 3 , Figure 3 This is another flowchart illustrating the BIM model standard layer identification method based on planar projection comparison in the embodiments of this application; Therefore, after step S206, the following steps are also included: S301. If there is at least one pair of different image units, then the different image units are recorded as difference image units. It should be noted that in the previously mentioned embodiments, if at least one pair of image units are different, it is directly identified as a non-standard layer. However, in this embodiment, further analysis will be initiated to determine whether it belongs to a non-standard layer.

[0063] S302. For each differential image unit, backtrack to the corresponding original three-dimensional space range and obtain the original three-dimensional space components within the space range. Backtracking refers to the operation of tracing the spatial location of a difference image unit in the original 3D floor model back to its original location based on the position information of the difference image unit in the 2D projection image. The original 3D spatial range refers to the actual area occupied by the part of the building component corresponding to the difference image unit in the original BIM model.

[0064] Since it is impossible to accurately determine whether a group of floor models with differences is a special standard floor by comparing only two-dimensional image units, it is necessary to go from the two-dimensional image level to the original three-dimensional space level, find the corresponding spatial range and components, and further explore the deeper structural information.

[0065] S303. Perform three-dimensional geometric similarity analysis on the three-dimensional spatial components corresponding to the difference image units; Three-dimensional geometric similarity analysis refers to the process of determining the degree of similarity between original three-dimensional spatial components corresponding to different image units from multiple dimensions such as geometric shape, spatial position relationship, and size ratio. Its purpose is to explore whether these components share the same structural features or specific proportional relationships in three-dimensional space, thereby determining whether the corresponding floor model group belongs to a special standard floor.

[0066] Specifically, for each set of original three-dimensional spatial components corresponding to a set of difference image units, the key geometric feature information of each component is first extracted. Based on methods such as size ratio calculation (comparing whether the ratio between the corresponding sizes of different components is constant) and spatial position matching (checking whether the relative positions between components conform to certain rules), these components are compared pairwise or analyzed as a whole to calculate the degree of geometric similarity between them and obtain the corresponding analysis results, such as similarity score, whether specific geometric relationships are satisfied, etc., in order to determine whether these components have the same geometric features or have proportional relationships.

[0067] S304. When three-dimensional spatial components are geometrically identical or proportionally related, they are determined to be standard layers with variable cross-sections. In some embodiments, step S304 specifically includes: When three-dimensional spatial components have the same geometric topology and their dimensional parameters have a constant scaling ratio in the horizontal or vertical direction, they are determined to be a variable cross-section standard layer relationship.

[0068] Among them, a standard floor with variable cross-section refers to a floor in a building where, although the corresponding floor model may show some differences when vertically projected and compared with image units, the original three-dimensional spatial components corresponding to the different image units are analyzed and found to have the same basic structural features in geometry. The only difference is that there is a certain scaling ratio in size (i.e., the size changes by a fixed ratio in the horizontal or vertical direction). Such a floor can still be identified as a standard floor, which is a special case of standard floor and is different from the conventional completely identical standard floor.

[0069] Having the same geometric topology means that the components are consistent in terms of spatial connection methods, hole distribution, and other topological features. It is an important basis for judging the similarity of components from the abstract spatial structure level.

[0070] The existence of a constant scaling ratio in the horizontal or vertical direction of dimensional parameters means that there is a fixed multiple relationship between the length, width, height and other dimensional values ​​of a component in the horizontal direction (such as along the transverse direction of the building) or the vertical direction (such as along the floor height direction of the building). For example, the horizontal length of a certain beam component is always scaled by a factor of 2 in different floors. This is used to define special geometric similarity cases.

[0071] Specifically, after obtaining the 3D geometric similarity analysis results of the 3D spatial components, for each set of components corresponding to the difference image units, the first step is to check whether their geometric topology is the same. This involves checking whether the spatial connection forms, internal holes, and other topological features of the components are consistent. For example, determining whether column components are all solid cylinders or hollow columns with the same number and location of holes, etc., establishes a basis for similarity from the overall spatial structure level. Then, the focus is on analyzing the dimensional parameters of the components. This involves examining whether there is a constant scaling ratio between the dimensional values ​​of each component in both the horizontal direction (e.g., the transverse span of the building, corresponding to the length, width, etc.) and the vertical direction (e.g., the floor height of the building, corresponding to the height, etc.). This can be achieved by calculating the ratios between the corresponding dimensions of different components to see if all ratios are close to a fixed constant (within the allowable error range). For example, for beam components across multiple floors, if the dimensional ratios in the length direction are consistently around 1.5, it indicates the existence of a constant scaling ratio in the horizontal direction. If analysis reveals that the components have the same geometric topology and a constant scaling ratio in the horizontal or vertical direction, then the floor model can be determined to be a standard floor with variable cross-section, and it can be included in the category of standard floors for further processing.

[0072] It is evident that the rule of determining a standard layer with variable cross-section when three-dimensional spatial components are geometrically identical or proportionally related has been clarified. Furthermore, the rule has been refined to the point that when three-dimensional spatial components have the same geometric topology and their dimensional parameters have a constant scaling ratio in the horizontal or vertical direction, they are determined to be a standard layer with variable cross-section. This is more in line with the actual situation and provides a clear and quantifiable standard for judgment.

[0073] S305. When the geometric proportions of three-dimensional spatial components are not the same or do not show a proportional relationship, they are judged as non-standard layers.

[0074] S306. Generate and output the standard layer recognition result containing standard layer information.

[0075] As can be seen, when determining whether a floor model is a standard floor, if at least one pair of image units are different, the corresponding original 3D spatial range is located back to obtain the original 3D spatial components within that range, allowing the analysis to move beyond surface image differences. Next, a 3D geometric similarity analysis is performed on the 3D spatial components corresponding to the differing image units. In practical use, some floor model groups may appear different after vertical projection, but they are actually special standard floors. Through geometric similarity analysis of the components, when the 3D spatial components are geometrically identical or proportionally related, they are identified as standard floors with variable cross-sections. This overcomes the limitations of relying solely on image comparison. Even if differences occur during the image comparison stage, subsequent 3D component analysis can uncover those special standard floor cases, avoiding the problem of misclassifying special standard floors as non-standard floors and improving the reliability of standard floor application throughout the entire cycle.

[0076] In practical applications, we often encounter situations where certain floors, which should rightfully be classified as standard floors, exhibit local variations. For example, a column running through multiple floors might have steps at both ends. When comparing image units at the same mesh location across different floor models using standard procedures, the steps at the top and bottom of the column, due to their distinct appearance, lead to these floors being misclassified as non-standard. Even subsequent 3D geometric similarity analysis, intended to correct these misclassifications through deeper component-level analysis, unfortunately fails to accurately identify these reasonable local differences due to the varying geometric features of the column's different parts caused by the steps, thus incorrectly excluding these floors that should belong to the standard floor category. More importantly, the current standard floor determination logic does not fully consider reasonable local variations such as columns and steps. This makes the scope of determination too narrow and easily misses floors that are actually standard floors but have reasonable local variations, thus affecting the accuracy and comprehensiveness of the standard floor determination results.

[0077] Therefore, in some embodiments, step S204 is replaced by: A1. Identify the projection outline of the floor projection image corresponding to the preset key component type, and delineate the component feature area on the two-dimensional floor projection image according to the position and range of the projection outline, and regard the component feature area as an image unit. Entity definitions: Preset key component types refer to pre-defined categories of building components in the architectural model that are of significant reference value for determining standard floors, such as columns, beams, and floor slabs. The characteristics of these components often play a crucial role in determining the structural consistency of the floors. Projected contour refers to the outer contour shape of the aforementioned key components on the floor projection image after projection. Component feature region refers to a specific area delineated on the two-dimensional floor projection image based on the projected contour of the key component. This region emphasizes the relevant characteristics of the corresponding key component and is subsequently used as a special image unit for analysis and comparison.

[0078] Specifically, firstly, based on pre-defined key component types (such as columns and beams to be the focus), image recognition technology (which can be machine learning-based image recognition algorithms or traditional methods such as geometric shape matching) is used to identify the projected outlines of these key components in the existing floor projection images. For example, for columns, the circular, rectangular, or other outer contour shapes and their coordinates in the 2D projection image are identified. Then, based on the position and extent of the identified projection outlines, the corresponding component feature regions are accurately delineated on the 2D floor projection image. This region can include the entire projection outline and a certain buffer range around it (such as extending outward by a few pixels, which can be set according to the actual component characteristics and analysis accuracy requirements). This comprehensively covers the main feature information of the key components. Finally, these delineated component feature regions are treated as special image units, distinct from ordinary image units obtained through conventional mesh cutting, for more targeted analysis later.

[0079] A2. Outside the component feature area, cut it into several image units according to a preset grid; Here, "outside the component feature area" refers to the remaining image portion in the two-dimensional floor projection image after removing the component feature areas of the corresponding key components defined in step A1.

[0080] It should be noted that outside the component feature area, the operation can be performed according to step S204, which will not be elaborated here.

[0081] Following step S305, the following is also included: A3. If the pixels between the feature regions of a component do not correspond completely, extract the distribution pattern of the difference pixels. Among them, the distribution pattern of differential pixels refers to the overall distribution law and shape of these different pixels in the feature area of ​​the component, such as being concentrated in a certain local area or presenting a certain geometric shape. By analyzing its distribution pattern, we can further explore the essential reasons for the differences and whether there are reasonable variations.

[0082] Specifically, after comparing the corresponding component feature regions in the projected images of different floors (by comparing pixels one by one or by calculating image feature similarity to determine if differences exist), once a case of incomplete pixel correspondence is found, the extraction and analysis of the distribution pattern of the differing pixels is initiated. By traversing each pixel within the component feature region, those differing pixels that are different from the corresponding pixels on other floors are marked (this can be determined by setting a pixel value difference threshold, such as color value differences exceeding a certain value). Then, the coordinate positions and quantities of these differing pixels within the component feature region are statistically analyzed, and their overall distribution patterns are examined. For example, are they concentrated at the edges of the component, in a corner, or distributed along a line? Furthermore, it is observed whether they exhibit a certain geometric shape (such as a circle or rectangle) in their arrangement trend. This process extracts the distribution pattern of the differing pixels, providing a basis for subsequent judgment of their rationality.

[0083] A4. Determine whether the distribution pattern constitutes a preset rule of geometric shape or outline that is the same as the component feature area. Among them, the preset geometric shapes or contours refer to some common graphic patterns with specific geometric shapes (such as hollow circles, rectangles, triangles, etc.) and specific contour features that are set in advance based on reasonable local variations in the building or relevant requirements for standard floor determination. These patterns are used to compare and judge the actual extracted difference pixel distribution patterns.

[0084] Specifically, after obtaining the distribution pattern information of the difference pixels extracted in step A3, it is compared and analyzed with the pre-defined geometric shapes or contours. First, the specific features and parameters of the pre-defined geometric shapes or contours are defined (for example, for a circle, its radius range, center position range, etc.; for a rectangle, the range of length and width, and angles, etc.). Then, it is checked whether the distribution pattern of the difference pixels can be fitted to one of these pre-defined geometric shapes or contours. This can be determined by mathematical fitting algorithms (such as least squares fitting of geometric shapes such as circles and rectangles) or by shape similarity judgment methods (such as calculating feature descriptors such as shape context to compare similarity). At the same time, it is also necessary to consider the correlation between this distribution pattern and the entire component feature area, that is, to determine whether it is in a reasonable position within the component feature area and whether it conforms to the structural and functional logic of the component itself (for example, for a column component, if the distribution pattern of the difference pixels forms a reasonable shape similar to a step at the top of the column and conforms to the functional logic of the column such as load-bearing). If the distribution pattern of the difference pixels can form a preset regular geometric shape or outline and is consistent with the overall features of the component feature area, then it is preliminarily determined that the relevant requirements of the standard layer are met at the component feature area level.

[0085] A5. If they are the same, change the non-standard layer to the standard layer.

[0086] The above embodiments present a problem that requires attention. In the original embodiments, the segmented regions were uniform, a feature that allowed for a one-to-one correspondence between image units in step S204. However, when step S204 is replaced, the replaced step first delineates the component feature regions around the projection outline of a preset key component type in the floor projection image and uses these as image units. Then, image units are segmented outside the component feature regions according to a preset grid. As a result, the component feature regions are likely to differ. Once the component feature regions differ, a chain reaction occurs, leading to inconsistent segmented regions. This inconsistency is directly reflected in the image units, meaning that the individual image units are no longer identical and cannot form an effective correspondence. This greatly complicates subsequent comparison work, making the comparison impossible to perform normally and ultimately resulting in a comparison failure, affecting the accuracy and reliability of the final result.

[0087] Therefore, in some embodiments, after step S204, the method further includes: A6, when there is a first image unit at the current grid position of the first floor model, but no corresponding second image unit at the current grid position of the second floor model, no comparison operation is performed, the first floor model is any floor model, the second floor model is any floor model in the same floor model group as the first floor model, the first image unit is any image unit, and the current grid position is any grid position.

[0088] Specifically, first, identify the current group of floor models involved in the analysis. Randomly select one floor model as the first floor model, and then randomly select another from the remaining floor models in the group as the second floor model (the selection order and specific model chosen do not affect the application of the rules; they are only used to determine the two objects for comparison). Then, for the already meshed floor projection images, traverse each mesh position (you can check each mesh position sequentially, starting from the top left corner and checking row by row, or use other reasonable traversal order). When a first image unit is found at a certain current mesh position in the first floor model, immediately check if a corresponding second image unit exists at the same current mesh position in the second floor model (that is, the mesh position with the same coordinates or row and column numbers in the projection images of the two floor models). If no corresponding image unit is found at that position in the second floor model, then skip the conventional comparison operations (such as skipping the comparison steps of calculating pixel similarity and feature differences of image units) according to the established rules.

[0089] A7. Calculate the spatial overlap area of ​​the third and fourth image units, which are located in the same grid position but have different sizes, on the two-dimensional floor projection image. Specifically, first, the two-dimensional floor projection images corresponding to the two floor models to be analyzed are determined. Then, based on the previous grid division rules, the third and fourth image units, which are in the same grid position but have different sizes, are found. These two image units can be regarded as graphics with a certain shape and range on the two-dimensional plane (such as rectangles, irregular polygons, etc., depending on the actual shape of the image unit). Next, their spatial overlap area is determined by the image coordinate system and geometric calculation methods. If the shape of the image unit is relatively regular (taking a rectangle as an example), the coordinate range of the overlapping part can be determined by comparing their coordinate ranges (such as the coordinates of the upper left vertex and the lower right vertex). That is, the horizontal coordinate is taken as the intersection of the horizontal coordinate ranges of the two image units, and the vertical coordinate is taken as the intersection, thus determining the rectangular range of the overlapping area. If the shape of the image unit is irregular, a polygon intersection algorithm (such as the scan line algorithm) can be used to treat the outline boundary of the image unit as a polygon. The spatial overlap area is determined by calculating the intersection between the polygons. Finally, the spatial overlap area information of these two image units on the two-dimensional floor projection image is obtained, which prepares for subsequent steps.

[0090] A8. Extract the third sub-image and the fourth sub-image that completely correspond to the spatially overlapping area from the third image unit and the fourth image unit, respectively. Specifically, based on the spatial overlap region information obtained in step A7 (such as the coordinate range and contour coordinate set of the overlap region), the operation is performed on the third image unit. If the third image unit is a regular shape (such as a rectangle), and the overlap region is also rectangular, the portion of the third image unit that matches the coordinate range of the overlap region is extracted through coordinate calculation and used as the third sub-image. If the third image unit is an irregular shape, the pixel set falling within the overlap region is extracted using an image segmentation algorithm (which can be based on pixel point judgment, extracting internal pixels along the contour boundary) to form the third sub-image, ensuring that the third sub-image exactly covers the spatial overlap region and contains the original image information within that region. The same method is used to operate on the fourth image unit to extract the fourth sub-image that completely corresponds to the spatial overlap region, preparing an accurate analysis object for subsequent comparison steps.

[0091] A9. Perform pairwise comparisons between the third and fourth sub-images.

[0092] The following describes an exemplary BIM model standard layer identification system 400 based on planar projection comparison provided in an embodiment of this application. Figure 4This is an exemplary hardware structure diagram of the BIM model standard layer recognition system 400 based on planar projection comparison provided in this application embodiment.

[0093] In some embodiments, the BIM model standard layer identification system 400 based on planar projection comparison is a computer device or includes a computer device in the BIM model standard layer identification system 400 based on planar projection comparison. The computer device includes a processor, memory, and a network interface connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores data. The network interface of the computer device is used to communicate with other external terminals or servers via a network connection. In some embodiments, the network interface can be a wired network interface; in some embodiments, the network interface can also be a wireless network interface. When the computer program is executed by the processor, it implements the methods in the embodiments of this application.

[0094] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0095] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0096] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0097] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0098] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for identifying standard layers in a BIM model based on planar projection comparison, characterized in that, include: Obtain the BIM model to be analyzed, and cut the BIM model according to the floor boundaries to obtain multiple floor models. The floor model is a three-dimensional sub-model corresponding to a single building floor. The floor models with the same floor height are grouped together to obtain several floor model groups; For each of the floor model groups, the floor models in the group are vertically projected in the same coordinate system to generate a corresponding two-dimensional floor projection image for each floor model. The floor projection image is cut into several image units according to a preset grid; Within the same group of floor models, the image units corresponding to different floor models and located at the same grid position are compared pairwise. If all corresponding image units in the floor model are identical, then the floor model is determined to be a standard floor.

2. The method according to claim 1, characterized in that, Following the step of determining that the floor model is a standard floor if all corresponding image units in the floor model are identical, the method further includes: If at least one pair of image units are different, the different image units are recorded as difference image units; For each of the differential image units, backtrack to the corresponding original three-dimensional spatial range, and obtain the original three-dimensional spatial components within the spatial range; Perform three-dimensional geometric similarity analysis on the three-dimensional spatial components corresponding to the difference image units; When the three-dimensional spatial components are geometrically identical or proportionally related, they are determined to be a standard layer with variable cross-section; Generate and output standard layer recognition results containing standard layer information.

3. The method according to claim 2, characterized in that, The step of determining a standard layer with variable cross-section when the three-dimensional spatial components are geometrically identical or proportionally related specifically includes: When three-dimensional spatial components have the same geometric topology and their dimensional parameters have a constant scaling ratio in the horizontal or vertical direction, they are determined to be a variable cross-section standard layer relationship.

4. The method according to claim 1, characterized in that, Before the step of grouping floor models with the same floor height into several floor model groups, the method further includes: All the floor models described are paired up in sequence to form corresponding floor groups; Compare the floor heights of the aforementioned floor groups; If the floor height of the current floor model is different from the floor height of all other floor models, then the current floor model is determined to be a non-standard floor. The current floor model is any of the floor models mentioned above, and the other floor models are the floor models excluding the current floor model.

5. The method according to claim 4, characterized in that, The step of grouping floor models with the same floor height into several floor model groups specifically includes: During the floor height comparison process, other floor models with the same floor height as the current floor model are collected and summarized; The collected and summarized other floor models, together with the current floor model, constitute the floor model group corresponding to the current floor model.

6. The method according to claim 1, characterized in that, Before the step of cutting the BIM model according to floor boundaries to obtain multiple floor models, wherein each floor model is a three-dimensional sub-model corresponding to a single building floor, the method further includes: The BIM model is converted into a 3D wireframe model.

7. The method according to claim 2, characterized in that, After the step of determining a standard layer with variable cross-section when the three-dimensional spatial components are geometrically identical or proportionally related, the method further includes: When the geometric proportions of the three-dimensional spatial components are not the same or not proportional, they are determined to be non-standard layers.

8. A BIM model standard layer identification system based on planar projection comparison, characterized in that, The BIM model standard layer identification system based on planar projection comparison includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the BIM model standard layer identification system based on planar projection comparison to perform the method as described in any one of claims 1-7.

9. A computer program product containing instructions, characterized in that, When the computer program product is run on the BIM model standard layer identification system based on planar projection comparison, the BIM model standard layer identification system based on planar projection comparison performs the method as described in any one of claims 1-7.

10. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is run on the BIM model standard layer identification system based on planar projection comparison, the BIM model standard layer identification system based on planar projection comparison performs the method as described in any one of claims 1-7.

Images