Data processing system and method for house acceptance

Abstract

The invention relates to the field of data processing, and particularly discloses a data processing system and method for house acceptance, and the system employs a data processing technology based on deep learning to carry out the comparative analysis of an acceptance BIM model and a design BIM model of a to-be-detected house mark, and extracts the house structure features of the acceptance BIM model and the design BIM model. And then space attention feature enhancement processing is carried out on the two, and feature difference measurement is carried out on the two on the basis, so that whether the to-be-detected house mark meets the acceptance standard of a construction party or not is intelligently determined. Therefore, traditional manual acceptance can be assisted or replaced through automatic data processing, so that the labor cost is reduced, and the efficiency and accuracy of house acceptance are improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and more specifically, to a data processing system and method for house acceptance. Background Art

[0002] With the rapid development of the construction industry and the continuous advancement of technological innovation, the design, construction and acceptance processes of buildings have become increasingly complex and changeable. The traditional house acceptance method mainly relies on manual verification and paper document comparison. This method is not only time-consuming and labor-intensive, but also easily affected by human factors, making it difficult to ensure the accuracy and credibility of the acceptance results. Especially when facing large-scale or complex construction projects, traditional methods often seem to be stretched and difficult to complete the acceptance tasks efficiently.

[0003] In recent years, with the rapid development of information technology, it has become possible to improve the efficiency and quality of house acceptance with the help of digital means. BIM technology, as a digital model that integrates various relevant information of construction projects, covers geometric information, spatial relationships, geographic information, and the properties and quantities of various building components. It can not only effectively improve the efficiency of building design and construction, but also make the operation and maintenance of buildings more efficient and convenient.

[0004] In the housing acceptance stage, although BIM technology can provide a comprehensive, accurate and real-time building information display platform, thereby greatly improving the accuracy and efficiency of the acceptance work, it is still necessary to rely on manual means to compare and judge the data information in the BIM model to ensure the accuracy of the acceptance results. This is not only a huge workload, but also prone to omissions.

[0005] Therefore, a data processing system and method for house acceptance is expected to be able to use the BIM model for automated acceptance. Summary of the invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a data processing system and method for house acceptance, which uses deep learning-based data processing technology to compare and analyze the acceptance BIM model and the design BIM model of the house subject to be tested, and extracts the house structure features of the acceptance BIM model and the design BIM model respectively, and then performs spatial attention feature enhancement processing on the two, and on this basis, measures the feature difference between the two, so as to intelligently determine whether the house subject to be tested meets the acceptance standards of the construction party. In this way, automated data processing can be used to assist or replace traditional manual acceptance, thereby reducing labor costs and improving the efficiency and accuracy of house acceptance.

[0007] According to one aspect of the present application, a data processing method for house acceptance is provided, which includes:

[0008] Obtaining the acceptance BIM model of the house subject to be tested, and extracting the design BIM model of the house subject to be tested from the database;

[0009] Extracting the house structure semantic features of the acceptance BIM model and the design BIM model to obtain a structural semantic coding feature graph of the acceptance BIM model and a structural semantic coding feature graph of the design BIM model;

[0010] Performing attention enhancement on the acceptance BIM model structure semantic coding feature graph and the design BIM model structure semantic coding feature graph to obtain an acceptance BIM model structure semantic space enhanced coding feature graph and a design BIM model structure semantic space enhanced coding feature graph;

[0011] Based on the feature difference between the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map, it is determined whether the house target to be tested meets the construction party's acceptance standards.

[0012] In the above-mentioned data processing method for house acceptance, attention enhancement is performed on the acceptance BIM model structure semantic coding feature map, including: based on the pixel spatial position information of the acceptance BIM model structure semantic coding feature map, adaptively attenuating and modulating the pixel-level semantic association structure of the acceptance BIM model structure semantic coding feature map to obtain the acceptance BIM model structure feature space attenuation pixel granularity semantic association weight map; based on the acceptance BIM model structure feature space attenuation pixel granularity semantic association weight map, feature modulating the acceptance BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space enhanced coding feature map.

[0013] In the above-mentioned data processing method for house acceptance, the house structure semantic features of the acceptance BIM model and the design BIM model are extracted to obtain the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map, including: inputting the acceptance BIM model and the design BIM model into a BIM model twin detection network including a first house model feature extractor and a second house model feature extractor to obtain the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map.

[0014] In the above-mentioned data processing method for house acceptance, based on the pixel spatial position information of the acceptance BIM model structure semantic coding feature map, the pixel-level semantic association structure of the acceptance BIM model structure semantic coding feature map is adaptively attenuated and modulated to obtain the acceptance BIM model structure feature space attenuation pixel granularity semantic association weight map, including: performing feature fine-grained spatial decoupling along the channel dimension on the acceptance BIM model structure semantic coding feature map to obtain a set of acceptance BIM model pixel granularity structure semantic feature vectors; any two acceptance BIM model pixel granularity structure semantic feature vectors in the set of acceptance BIM model pixel granularity structure semantic feature vectors are input into the semantic An association score measurement network is used to obtain a set of semantic association score vectors between pixel granularities of the acceptance BIM model; based on the spatial span between any two pixel granularity structural semantic feature vectors of the acceptance BIM model in the set of pixel granularity structural semantic feature vectors of the acceptance BIM model, the set of semantic association score vectors between pixel granularities of the acceptance BIM model is subjected to semantic association adaptive attenuation modulation to obtain a set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model; the set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model is subjected to dimensional reconstruction and weight allocation to obtain a semantic association weight map between spatial attenuation pixel granularities of the acceptance BIM model structural feature.

[0015] In the above-mentioned data processing method for house acceptance, any two of the acceptance BIM model pixel granularity structure semantic feature vectors in the set of the acceptance BIM model pixel granularity structure semantic feature vectors are input into the semantic association score measurement network to obtain a set of semantic association score vectors between pixel granularities of the acceptance BIM model, including: cascading and fusing any two of the acceptance BIM model pixel granularity structure semantic feature vectors in the set of the acceptance BIM model pixel granularity structure semantic feature vectors, multiplying them by a weight parameter matrix, and then performing dot addition with a bias vector to obtain a set of semantic association score vectors between pixel granularities of the acceptance BIM model.

[0016] In the above-mentioned data processing method for house acceptance, based on the spatial span between any two of the acceptance BIM model pixel granularity structure semantic feature vectors in the set of the acceptance BIM model pixel granularity structure semantic feature vectors, a set of semantic association score vectors between pixel granularities of the acceptance BIM model is subjected to semantic association adaptive attenuation modulation to obtain a set of acceptance BIM model spatial attenuation pixel granularity semantic association score vectors, including: spatially annotating each acceptance BIM model pixel granularity structure semantic feature vector in the set of the acceptance BIM model pixel granularity structure semantic feature vectors to obtain a set of acceptance BIM model pixel spatial position data; calculating the acceptance BIM model pixel spatial attenuation ... The Euclidean distance between the spatial position data of any two acceptance BIM model pixel granularity structure semantic feature vectors in the set of space position data is used as the spatial span value of the two acceptance BIM model pixel granularity structure semantic feature vectors to obtain a set of acceptance BIM model pixel granularity spatial span values; based on each acceptance BIM model pixel granularity spatial span value in the set of acceptance BIM model pixel granularity spatial span values, each acceptance BIM model pixel granularity semantic association score vector in the set of acceptance BIM model pixel granularity semantic association score vectors is subjected to semantic association adaptive attenuation modulation to obtain a set of acceptance BIM model spatial attenuation pixel granularity semantic association score vectors.

[0017] In the above-mentioned data processing method for house acceptance, based on each acceptance BIM model pixel granularity spatial span value in the set of the acceptance BIM model pixel granularity spatial span value, each acceptance BIM model pixel granularity semantic association score vector in the set of the acceptance BIM model pixel granularity semantic association score vector is subjected to semantic association adaptive attenuation modulation to obtain the set of the acceptance BIM model spatial attenuation pixel granularity semantic association score vectors, including: dividing the acceptance BIM model pixel granularity spatial span value by a preset scaling factor to obtain the acceptance BIM model pixel granularity spatial span scaling value; calculating an exponential function value with e as the base and the opposite number of the acceptance BIM model pixel granularity spatial span scaling value as the exponent to obtain a semantic association spatial attenuation coefficient; calculating the acceptance BIM model pixel granularity semantic association score vector divided by the semantic association spatial attenuation coefficient to obtain the acceptance BIM model spatial attenuation pixel granularity semantic association score vector.

[0018] In the above-mentioned data processing method for house acceptance, the set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model is dimensionally reconstructed and weighted to obtain the semantic association weight map between spatial attenuation pixel granularities of the acceptance BIM model structural features, including: arranging the set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model into a semantic association score feature map between spatial attenuation pixel granularities of the acceptance BIM model, and then inputting it into a feature dimension modulation layer based on a point convolution layer and a weight allocation network based on a Sigmoid function to obtain the semantic association weight map between spatial attenuation pixel granularities of the acceptance BIM model structural features.

[0019] In the above-mentioned data processing method for house acceptance, feature modulation is performed on the acceptance BIM model structure semantic coding feature map based on the semantic association weight map between the attenuated pixel granularities of the acceptance BIM model structure feature space to obtain the acceptance BIM model structure semantic space enhanced coding feature map, including: calculating the position point multiplication between the semantic association weight map between the attenuated pixel granularities of the acceptance BIM model structure feature space and the acceptance BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space enhanced coding feature map.

[0020] In the above-mentioned data processing method for house acceptance, based on the feature difference between the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map, it is determined whether the house target to be tested meets the construction party's acceptance standards, including: inputting the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map into the BIM model difference measurement network to obtain the acceptance-design house target difference semantic coding feature map; inputting the acceptance-design house target difference semantic coding feature map into the classifier-based intelligent auxiliary acceptance module to obtain an auxiliary acceptance result, and the auxiliary acceptance result is used to indicate whether it meets the construction party's acceptance standards.

[0021] According to another aspect of the present application, a data processing system for house acceptance is provided, comprising:

[0022] A BIM model acquisition module is used to acquire the acceptance BIM model of the house subject to be tested, and extract the design BIM model of the house subject to be tested from the database;

[0023] A housing structure semantic feature extraction module is used to extract the housing structure semantic features of the acceptance BIM model and the design BIM model to obtain a structural semantic coding feature graph of the acceptance BIM model and a structural semantic coding feature graph of the design BIM model;

[0024] An attention enhancement module is used to perform attention enhancement on the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map;

[0025] The acceptance standard evaluation module is used to determine whether the house subject to be tested meets the acceptance standard of the construction party based on the feature difference between the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map.

[0026] Compared with the prior art, the data processing system and method for housing acceptance provided by the present application uses a data processing technology based on deep learning to compare and analyze the acceptance BIM model and the design BIM model of the housing target to be tested, respectively extract the housing structure features of the acceptance BIM model and the design BIM model, and then perform spatial attention feature enhancement processing on the two, and on this basis, measure the feature difference between the two, so as to intelligently determine whether the housing target to be tested meets the acceptance standards of the construction party. In this way, automated data processing can be used to assist or replace traditional manual acceptance, thereby reducing labor costs and improving the efficiency and accuracy of housing acceptance. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0028] Figure 1 The present invention is a flowchart of a data processing method for house acceptance according to an embodiment of the present application.

[0029] Figure 2 A data flow diagram of a data processing method for house acceptance according to an embodiment of the present application.

[0030] Figure 3 This is a flowchart of sub-step S3 of the data processing method for house acceptance according to an embodiment of the present application.

[0031] Figure 4 This is a flowchart of sub-step S31 of the data processing method for house acceptance according to an embodiment of the present application.

[0032] Figure 5 This is a flowchart of sub-step S313 of the data processing method for house acceptance according to an embodiment of the present application.

[0033] Figure 6 It is a block diagram of a data processing system for house acceptance according to an embodiment of the present application. DETAILED DESCRIPTION

[0034] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0035] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0036] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0037] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0038] In response to the technical problems described in the above background technology, this application proposes an optimized data processing method for house acceptance, which uses deep learning-based data processing technology to compare and analyze the acceptance BIM model and design BIM model of the house subject to be tested, and extracts the house structure features of the acceptance BIM model and the design BIM model respectively, and then performs spatial attention feature enhancement processing on the two, and on this basis, measures the feature difference between the two, so as to intelligently determine whether the house subject to be tested meets the acceptance standards of the construction party. In this way, automated data processing can be used to assist or replace traditional manual acceptance, thereby reducing labor costs and improving the efficiency and accuracy of house acceptance.

[0039] Figure 1 The present invention is a flowchart of a data processing method for house acceptance according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the data processing method for house acceptance according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the data processing method for house acceptance includes the following steps: S1, obtaining the acceptance BIM model of the house subject to be tested, and extracting the design BIM model of the house subject to be tested from the database; S2, extracting the house structure semantic features of the acceptance BIM model and the design BIM model to obtain the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map; S3, performing attention enhancement on the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space reinforcement coding feature map and the design BIM model structure semantic space reinforcement coding feature map; S4, determining whether the house subject to be tested meets the construction party's acceptance standard based on the feature difference between the acceptance BIM model structure semantic space reinforcement coding feature map and the design BIM model structure semantic space reinforcement coding feature map.

[0040] In the above-mentioned data processing method for house acceptance, the step S1 obtains the acceptance BIM model of the house subject to be tested, and extracts the design BIM model of the house subject to be tested from the database. It should be understood that the design BIM model represents the original design intention and planning of the project, and contains detailed information such as the expected building structure, layout, and material use. The acceptance BIM model is a digital model constructed based on the house after the actual construction is completed, reflecting the results of the actual construction. By comparing and analyzing the two, any deviations or changes that occur during the construction process can be intuitively discovered, thereby accurately evaluating whether the house meets the design requirements and construction standards.

[0041] In the above-mentioned data processing method for house acceptance, the step S2 extracts the house structure semantic features of the acceptance BIM model and the design BIM model to obtain the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map. In a specific example of the present application, the step S2 further includes: inputting the acceptance BIM model and the design BIM model into a BIM model twin detection network including a first house model feature extractor and a second house model feature extractor to obtain the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map. It should be understood that, considering that the original BIM model data volume is large and complex, direct comparison calculation is inefficient and difficult to process. Therefore, in order to reduce the calculation complexity and extract the key structural features in the BIM model, the present application further uses the BIM model twin detection network to process the acceptance BIM model and the design BIM model. The BIM model twin detection network includes a weight-shared first house model feature extractor and a second house model feature extractor, both of which are based on a three-dimensional convolutional neural network architecture and are used to extract structured information such as the position, size and attributes of building elements in the acceptance BIM model and the design BIM model, and generate corresponding acceptance BIM model structure semantic coding feature maps and design BIM model structure semantic coding feature maps. In addition, through this weight-shared twin network architecture, the consistency of feature extraction of the acceptance BIM model and the design BIM model can be ensured, thereby improving the reliability and accuracy of subsequent feature comparison analysis.

[0042] In the above-mentioned data processing method for house acceptance, the step S3 performs attention enhancement on the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map. It should be understood that considering the different importance of different building elements in the BIM model for house acceptance, for example, key nodes in the structure and important building components are usually more critical than non-structural decorative elements. Therefore, in order to improve the model's ability to capture key structural patterns in the BIM model, the present application proposes an adaptive feature enhancement method based on spatial context perception, which dynamically adjusts its feature weights by learning the spatial distribution and contextual semantic association structure of different building elements in the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map, and adaptively focuses on key structural features of the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map, thereby improving the feature discrimination and the model's sensitivity to house structure differences. Among them, Figure 3FIG. 4 is a flowchart of sub-step S3 of the data processing method for house acceptance according to an embodiment of the present application. Figure 3 As shown, the step S3 includes the steps of: S31, based on the pixel spatial position information of the acceptance BIM model structure semantic coding feature map, adaptively attenuating and modulating the pixel-level semantic association structure of the acceptance BIM model structure semantic coding feature map to obtain the acceptance BIM model structure feature space attenuation pixel granularity semantic association weight map; S32, based on the acceptance BIM model structure feature space attenuation pixel granularity semantic association weight map, feature modulating the acceptance BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space enhanced coding feature map.

[0043] Figure 4 FIG. 4 is a flowchart of sub-step S31 of the data processing method for house acceptance according to an embodiment of the present application. Figure 4 As shown, the step S31 includes the steps of: S311, performing feature fine-grained spatial decoupling along the channel dimension on the acceptance BIM model structure semantic encoding feature map to obtain a set of pixel granularity structure semantic feature vectors of the acceptance BIM model; S312, inputting any two pixel granularity structure semantic feature vectors of the acceptance BIM model in the set of pixel granularity structure semantic feature vectors of the acceptance BIM model into a semantic association score measurement network to obtain a set of semantic association score vectors between pixel granularities of the acceptance BIM model; S313, performing semantic association adaptive attenuation modulation on the set of semantic association score vectors between pixel granularities of the acceptance BIM model based on the spatial span between any two pixel granularity structure semantic feature vectors of the acceptance BIM model in the set of pixel granularity structure semantic feature vectors of the acceptance BIM model to obtain a set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model; S314, performing dimension reconstruction and weight allocation on the set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model to obtain a semantic association weight map between spatial attenuation pixel granularities of the structural feature of the acceptance BIM model.

[0044] More specifically, the step S312 further includes: cascading and fusing any two of the acceptance BIM model pixel granularity structure semantic feature vectors in the set of the acceptance BIM model pixel granularity structure semantic feature vectors, multiplying them by a weight parameter matrix, and then performing dot addition with a bias vector to obtain a set of semantic association score vectors between pixel granularities of the acceptance BIM model.

[0045] That is, firstly, the fine-grained spatial decoupling of the semantic encoding feature map of the acceptance BIM model structure is performed along the channel dimension, and it is decomposed into feature vectors of multiple pixel granularities, thereby enhancing the model's recognition ability of local detail features. Then, considering that the local features of objects of the same type usually have stronger semantic associations, the present application further uses the trained network model to perform semantic association measurement on any two pixel granularity structural semantic feature vectors of the acceptance BIM model in the set to estimate the semantic connection strength between the two, and generate a set of semantic association score vectors between pixel granularities of the acceptance BIM model, thereby revealing the semantic connection between different building elements in the acceptance BIM model.

[0046] Figure 5 FIG. 4 is a flowchart of sub-step S313 of the data processing method for house acceptance according to an embodiment of the present application. Figure 5 As shown, the step S313 includes the steps of: S3131, spatially annotating the spatial position of each acceptance BIM model pixel granularity structure semantic feature vector in the set of the acceptance BIM model pixel granularity structure semantic feature vector to obtain a set of acceptance BIM model pixel spatial position data; S3132, calculating the Euclidean distance between the spatial position data of any two acceptance BIM model pixel granularity structure semantic feature vectors in the set of the acceptance BIM model pixel spatial position data as the spatial span value of the two acceptance BIM model pixel granularity structure semantic feature vectors to obtain a set of acceptance BIM model pixel granularity spatial span values; S3133, based on each acceptance BIM model pixel granularity spatial span value in the set of the acceptance BIM model pixel granularity spatial span value, performing semantic association adaptive attenuation modulation on each acceptance BIM model pixel granularity semantic association score vector in the set of the acceptance BIM model pixel granularity semantic association score vector to obtain a set of acceptance BIM model spatial attenuation pixel granularity semantic association score vectors.

[0047] In particular, the present application considers that even if two local features present similar characteristic attributes, they may carry completely different information meanings if they are located far apart in space. Therefore, the present application further annotates the spatial position of the semantic feature vectors of the pixel granular structure of each acceptance BIM model, adds its exact position information in the original BIM model, and calculates the spatial span between any two semantic feature vectors of the pixel granular structure of the acceptance BIM model based on this, thereby revealing the relative position relationship between the local features of each pixel granularity, so as to filter out interference items that are similar on the surface but not actually relevant in the subsequent feature modulation process.

[0048] In a specific example of the present application, the step S3133 further includes: dividing the acceptance BIM model pixel granularity spatial span value by a preset scaling factor to obtain the acceptance BIM model pixel granularity spatial span scaling value; calculating an exponential function value with e as the base and the opposite number of the acceptance BIM model pixel granularity spatial span scaling value as the exponent to obtain a semantic association spatial attenuation coefficient; calculating the semantic association score vector between pixel granularities of the acceptance BIM model divided by the semantic association spatial attenuation coefficient to obtain the acceptance BIM model spatial attenuation pixel granularity semantic association score vector. That is, based on the spatial span value between any two acceptance BIM model pixel granularity structure semantic feature vectors, the semantic association score vector between the two is adaptively attenuated and modulated. When two pixel granularity local features show strong semantic correlation and are close to each other, the importance of both will be maximized; on the contrary, if the two have high semantic similarity but are far apart, their influence will be appropriately weakened, thereby accurately reflecting the most noteworthy feature parts in the BIM model.

[0049] More specifically, the step S314 further includes: after arranging the set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model into a feature map of semantic association score between spatial attenuation pixel granularities of the acceptance BIM model, inputting it into a feature dimension modulation layer based on a point convolution layer and a weight allocation network based on a Sigmoid function to obtain a weight map of semantic association between spatial attenuation pixel granularities of the acceptance BIM model structure feature. That is, after arranging the set of semantic association score vectors between pixel granularities of the acceptance BIM model after spatial attenuation modulation into a feature map form, performing feature dimension modulation and weight allocation on it through point convolution and Sigmoid function, and generating a corresponding weight map of semantic association between spatial attenuation pixel granularities of the acceptance BIM model structure feature. It is worth mentioning that point convolution is an operation in a convolutional neural network, which can reduce or increase the depth of the feature map by changing the number of convolution kernels, which helps to extract features of different scales and enhance the expression ability of the model.

[0050] Specifically, the step S32 further includes: calculating the semantic association weight map between the attenuated pixel granularity of the acceptance BIM model structure feature space and the acceptance BIM model structure semantic coding feature map by position point multiplication to obtain the acceptance BIM model structure semantic space enhanced coding feature map. That is, the acceptance BIM model structure semantic coding feature map is weighted modulated by element-by-element multiplication operation, so as to selectively amplify important information and suppress irrelevant background information or non-critical feature areas, further highlighting the expression of key structural features in the acceptance BIM model, thereby obtaining the acceptance BIM model structure semantic space enhanced coding feature map.

[0051] Accordingly, the step S3 includes: processing the acceptance BIM model structure semantic coding feature map with the following feature enhancement formula to obtain the acceptance BIM model structure semantic space enhanced coding feature map, wherein the feature enhancement formula is:

[0052]

[0053]

[0054]

[0055]

[0056] A∈R (H×W)×(H×W)×C

[0057] A'=C pw (A); A'∈R H×W×C

[0058] Y=σ(A′)

[0059]

[0060] F'=F⊙Y

[0061] Where F is the semantic encoding feature graph of the acceptance BIM model structure, and F∈R H×W×C , where C, H and W are the number of channels, height and width of the semantic encoding feature map of the acceptance BIM model structure, respectively. decouple(·) represents the feature decoupling operation. X is the set of semantic feature vectors of the pixel granularity structure of the acceptance BIM model. v h,w is the semantic feature vector of pixel granularity structure of the acceptance BIM model, and C is the length of the semantic feature vector of pixel granularity structure of the acceptance BIM model, p h,w is the spatial position data of the semantic feature vector of the pixel granular structure of the acceptance BIM model, where p h,w =(h,w), (h,w) is the spatial position coordinate of the semantic feature vector of the pixel granularity structure of the acceptance BIM model, P represents the set of pixel spatial position data of the acceptance BIM model, represents any two pixel granularity structure semantic feature vectors of the acceptance BIM model in the set of pixel granularity structure semantic feature vectors of the acceptance BIM model, W r and b r Represent the weight parameter matrix and bias vector respectively, represents the semantic association score vector between pixel granularities of the acceptance BIM model, S is the set of semantic association score vectors between pixel granularities of the acceptance BIM model, is the spatial span value between any two pixel granularity structure semantic feature vectors of the acceptance BIM model, P' is the set of pixel granularity spatial span values ​​of the acceptance BIM model, is the preset scaling factor, is the semantic association score vector between the spatial attenuation pixel granularity of the acceptance BIM model, S' is the set of semantic association score vectors between the spatial attenuation pixel granularity of the acceptance BIM model, A represents the semantic association score feature map between the spatial attenuation pixel granularity of the acceptance BIM model, C pw (·) represents the point convolution operation, A' is the semantic association score feature map between spatial attenuated pixel granularities of the acceptance BIM model after feature dimension modulation, σ(·) is the Sigmoid function, Y represents the semantic association weight map between spatial attenuated pixel granularities of the structural feature space of the acceptance BIM model, ⊙ represents the point multiplication by position, and F' represents the structural semantic space enhanced coding feature map of the acceptance BIM model.

[0062] In particular, the present application processes the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map respectively through the above-mentioned processing method, which can effectively enhance the model's recognition and distinction capabilities for key structural features in house acceptance, and obtain the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map enhanced by spatial context awareness, thereby helping to achieve more refined feature comparison.

[0063] In the above-mentioned data processing method for house acceptance, the step S4 determines whether the house subject to be tested meets the acceptance criteria of the construction party based on the feature difference between the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map. In a specific example of the present application, the step S4 further includes: inputting the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map into the BIM model difference measurement network to obtain the difference semantic coding feature map of the acceptance-design house subject; inputting the difference semantic coding feature map of the acceptance-design house subject into the intelligent auxiliary acceptance module based on the classifier to obtain the auxiliary acceptance result, and the auxiliary acceptance result is used to indicate whether it meets the acceptance criteria of the construction party. It should be understood that in order to reveal the feature difference between the acceptance BIM model and the design BIM model, the present application uses the BIM model difference measurement network to compare and analyze the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map. In a specific example of the present application, the BIM model difference measurement network calculates the positional difference between the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map to generate a difference semantic coding feature map of the acceptance-design house mark, thereby intuitively reflecting the differences in structure and details between the acceptance BIM model and the design BIM model, and providing a more accurate reference for house acceptance. It is worth mentioning that the classifier is based on a multi-layer perceptron (MLP) structure. Through a large amount of training, it can effectively learn the feature differences between the acceptance BIM model and the design BIM model contained in the difference semantic coding feature map of the acceptance-design house mark, fully understand the similarity correlation between the two, and then accurately distinguish whether the house mark to be tested meets the construction party's acceptance standards, and output the corresponding acceptance results.

[0064] In a preferred example of the present application, inputting the differential semantic coding feature map of the acceptance-design house mark into an intelligent auxiliary acceptance module based on a classifier to obtain an auxiliary acceptance result includes: expanding the differential semantic coding feature map of the acceptance-design house mark into a differential semantic coding feature vector of the acceptance-design house mark; determining a differential semantic coding association response matrix of the acceptance-design house mark and a differential semantic coding distance response matrix of the acceptance-design house mark based on the association value and distance value of the eigenvalues ​​of the differential semantic coding feature vector of the acceptance-design house mark; matrix multiplying the differential semantic coding feature vector of the acceptance-design house mark with the differential semantic coding association response matrix of the acceptance-design house mark to obtain a differential semantic coding association response vector of the acceptance-design house mark; and multiplying the differential semantic coding distance response matrix of the acceptance-design house mark with the differential semantic coding feature vector of the acceptance-design house mark. The transposed vector of the code feature vector is matrix multiplied to obtain the difference semantic coding distance response vector of the acceptance-design house mark; the difference semantic coding association response matrix of the acceptance-design house mark and the difference semantic coding distance response matrix of the acceptance-design house mark are dot-multiplied, and then the matrix is ​​multiplied with the transposed vector of the difference semantic coding feature vector of the acceptance-design house mark to obtain the difference semantic coding logic bias vector of the acceptance-design house mark; the difference semantic coding association response vector of the acceptance-design house mark, the difference semantic coding distance response vector of the acceptance-design house mark and the difference semantic coding logic bias vector of the acceptance-design house mark are dot-multiplied to obtain the optimized difference semantic coding feature vector of the acceptance-design house mark; the optimized difference semantic coding feature vector of the acceptance-design house mark is input into the intelligent auxiliary acceptance module based on the classifier to obtain the auxiliary acceptance result.

[0065] Accordingly, the optimization process of the difference semantic coding feature vector of the acceptance-design house mark is expressed by the following optimization formula:

[0066] M 1 (i,j)=v i ×v j

[0067]

[0068] v i ,v j ∈V∈R 1×L

[0069] V 1 ∈R 1×L

[0070] V 2 ∈R L×1

[0071]

[0072] Among them, M 1 The differential semantic encoding association response matrix for acceptance-design house label, M 2 The distance response matrix for the difference semantic encoding of acceptance-design house label, M 1 (i, j) is the eigenvalue of the position (i, j) in the differential semantic encoding association response matrix of the acceptance-design house mark, M 2 (i, j) is the eigenvalue of the position (i, j) in the difference semantic encoding distance response matrix of the acceptance-design house mark, v i and v j are the i-th and j-th eigenvalues ​​in the differential semantic encoding feature vector of the acceptance-design house mark, V 1 The point-to-point vector between the difference semantic coding association response vector of the acceptance-design house mark and the difference semantic coding distance response vector of the acceptance-design house mark, V 2 is the logic bias vector of the difference semantic coding of the acceptance-design house mark, R is a real number set, L is the length of the difference semantic coding feature vector of the acceptance-design house mark, ⊙ represents the point multiplication by position, represents matrix multiplication, It represents addition by location point, V is the difference semantic coding feature vector of acceptance-design house mark, and V' is the difference semantic coding feature vector of optimized acceptance-design house mark.

[0073] That is, the present application takes into account that when the acceptance BIM model structural semantic space enhancement coding feature map and the design BIM model structural semantic space enhancement coding feature map respectively represent the model space enhancement semantic coding features of the acceptance BIM model of the house target to be tested and the design BIM model of the house target to be tested, when they are input into the BIM model difference measurement network, the measurement error introduced during modeling of the acceptance BIM model of the house target to be tested and the spatial distribution significance introduced when the feature attention of the model semantic coding features is enhanced will cause the BIM model difference features between the acceptance BIM model structural semantic space enhancement coding feature map and the design BIM model structural semantic space enhancement coding feature map to have large-scale simple repetition, which affects the logical dependency of the model semantic difference and reduces the accuracy of the classification results based on the distribution of model semantic differences.

[0074] Therefore, by using the self-correlation matrix and self-distance matrix of the differential semantic coding feature vector of the acceptance-design house mark after the differential semantic coding feature map of the acceptance-design house mark is expanded as the statistical reference-free distribution response framework of the differential semantic coding feature map of the acceptance-design house mark, based on the retrieval-enhanced reverse response construction of the differential semantic coding feature vector of the acceptance-design house mark, simple repetition of feature distribution is avoided, and by ensuring the intrinsic mapping logic of the differential semantic coding feature vector of the acceptance-design house mark based on the retrieval-response context, the surface combinatorial mapping of the differential semantic coding feature map of the acceptance-design house mark is avoided, so as to achieve the logical dependency mapping of the differential semantic coding feature map of the acceptance-design house mark to the classification target domain while maintaining the intuitive response correlation, thereby improving the accuracy of the auxiliary acceptance result obtained by inputting the differential semantic coding feature map of the acceptance-design house mark into the intelligent auxiliary acceptance module based on the classifier.

[0075] In summary, the data processing method for house acceptance based on the embodiment of the present application is explained, which uses a data processing technology based on deep learning to compare and analyze the acceptance BIM model and the design BIM model of the house subject to be tested, and extracts the house structure features of the acceptance BIM model and the design BIM model respectively, and then performs spatial attention feature enhancement processing on both, and on this basis, measures the feature difference between the two, so as to intelligently determine whether the house subject to be tested meets the acceptance criteria of the construction party. In this way, automated data processing can be used to assist or replace traditional manual acceptance, thereby reducing labor costs and improving the efficiency and accuracy of house acceptance.

[0076] Furthermore, a data processing system for house acceptance is also provided.

[0077] Figure 6 FIG. 1 is a block diagram of a data processing system for house inspection according to an embodiment of the present application. Figure 6As shown, according to the data processing system 100 for house acceptance of the embodiment of the present application, it includes: a BIM model acquisition module 110, which is used to acquire the acceptance BIM model of the house target to be tested, and extract the design BIM model of the house target to be tested from the database; a house structure semantic feature extraction module 120, which is used to extract the house structure semantic features of the acceptance BIM model and the design BIM model to obtain the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map; an attention enhancement module 130, which is used to enhance the attention of the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space enhancement coding feature map and the design BIM model structure semantic space enhancement coding feature map; an acceptance standard evaluation module 140, which is used to determine whether the house target to be tested meets the construction party's acceptance standard based on the feature difference between the acceptance BIM model structure semantic space enhancement coding feature map and the design BIM model structure semantic space enhancement coding feature map.

[0078] Here, those skilled in the art will appreciate that the specific operations of each module in the data processing system for house acceptance are already described in the above reference. Figures 1 to 5 The data processing method for house acceptance has been introduced in detail, and therefore, its repeated description will be omitted.

[0079] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.

[0080] In the above embodiments, the description of each embodiment has its own emphasis. For the parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0081] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. 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, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.

[0082] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0083] Finally, it should be noted that the above description has been given for the purpose of illustration and description. In addition, the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A data processing method for house acceptance, characterized in that: include: Obtaining the acceptance BIM model of the house subject to be tested, and extracting the design BIM model of the house subject to be tested from the database; Extracting the house structure semantic features of the acceptance BIM model and the design BIM model to obtain a structural semantic coding feature graph of the acceptance BIM model and a structural semantic coding feature graph of the design BIM model; Performing attention enhancement on the acceptance BIM model structure semantic coding feature graph and the design BIM model structure semantic coding feature graph to obtain an acceptance BIM model structure semantic space enhanced coding feature graph and a design BIM model structure semantic space enhanced coding feature graph; Based on the feature difference between the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map, determining whether the house subject to be tested meets the construction party's acceptance criteria; Among them, the attention enhancement is performed on the acceptance BIM model structure semantic coding feature map, including: based on the pixel spatial position information of the acceptance BIM model structure semantic coding feature map, adaptively attenuating and modulating the pixel-level semantic association structure of the acceptance BIM model structure semantic coding feature map to obtain the acceptance BIM model structure feature space attenuation pixel granularity semantic association weight map; based on the acceptance BIM model structure feature space attenuation pixel granularity semantic association weight map, feature modulating the acceptance BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space enhanced coding feature map.

2. The data processing method for housing acceptance according to claim 1, characterized in that: Extracting the house structure semantic features of the acceptance BIM model and the design BIM model to obtain the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map, including: The acceptance BIM model and the design BIM model are input into a BIM model twin detection network including a first house model feature extractor and a second house model feature extractor to obtain a structural semantic encoding feature graph of the acceptance BIM model and a structural semantic encoding feature graph of the design BIM model.

3. The data processing method for housing acceptance according to claim 2, characterized in that: Based on the pixel spatial position information of the acceptance BIM model structure semantic coding feature map, the pixel-level semantic association structure of the acceptance BIM model structure semantic coding feature map is adaptively attenuated and modulated to obtain a semantic association weight map between attenuated pixel granularities of the acceptance BIM model structure feature space, including: Performing feature fine-grained spatial decoupling along the channel dimension on the acceptance BIM model structure semantic encoding feature map to obtain a set of acceptance BIM model pixel granularity structure semantic feature vectors; Input any two of the acceptance BIM model pixel granularity structure semantic feature vectors in the set of the acceptance BIM model pixel granularity structure semantic feature vectors into a semantic association score measurement network to obtain a set of semantic association score vectors between pixel granularities of the acceptance BIM model; Based on the spatial span between any two semantic feature vectors of the pixel granularity structure of the acceptance BIM model in the set of semantic feature vectors of the pixel granularity structure of the acceptance BIM model, the set of semantic association score vectors between pixel granularities of the acceptance BIM model is subjected to semantic association adaptive attenuation modulation to obtain a set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model; The set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model is dimensionally reconstructed and weighted to obtain a semantic association weight map between spatial attenuation pixel granularities of the acceptance BIM model structural feature.

4. The data processing method for housing acceptance according to claim 3 is characterized in that: Inputting any two of the acceptance BIM model pixel granularity structure semantic feature vectors in the set of the acceptance BIM model pixel granularity structure semantic feature vectors into the semantic association score measurement network to obtain a set of semantic association score vectors between pixel granularities of the acceptance BIM model, including: Any two of the acceptance BIM model pixel granularity structure semantic feature vectors in the set of the acceptance BIM model pixel granularity structure semantic feature vectors are cascaded and fused, multiplied by a weight parameter matrix, and then dotted with a bias vector to obtain a set of semantic association score vectors between pixel granularities of the acceptance BIM model.

5. The data processing method for housing acceptance according to claim 4, characterized in that: Based on the spatial span between any two semantic feature vectors of the pixel granularity structure of the acceptance BIM model in the set of semantic feature vectors of the pixel granularity structure of the acceptance BIM model, a set of semantic association score vectors between pixel granularities of the acceptance BIM model is subjected to semantic association adaptive attenuation modulation to obtain a set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model, including: Performing spatial position annotation on each acceptance BIM model pixel granularity structure semantic feature vector in the set of acceptance BIM model pixel granularity structure semantic feature vectors to obtain a set of acceptance BIM model pixel spatial position data; Calculating the Euclidean distance between the spatial position data of any two pixel granularity structure semantic feature vectors of the acceptance BIM model in the set of pixel spatial position data of the acceptance BIM model as the spatial span value of the two pixel granularity structure semantic feature vectors of the acceptance BIM model to obtain a set of pixel granularity spatial span values ​​of the acceptance BIM model; Based on each acceptance BIM model pixel granularity spatial span value in the set of the acceptance BIM model pixel granularity spatial span values, each acceptance BIM model pixel granularity semantic association score vector in the set of the acceptance BIM model pixel granularity semantic association adaptive attenuation modulation is performed to obtain the set of the acceptance BIM model spatial attenuation pixel granularity semantic association score vectors.

6. The data processing method for housing inspection according to claim 5, characterized in that: Based on each acceptance BIM model pixel granularity spatial span value in the set of acceptance BIM model pixel granularity spatial span values, each acceptance BIM model pixel granularity semantic association score vector in the set of acceptance BIM model pixel granularity semantic association score vectors is subjected to semantic association adaptive attenuation modulation to obtain the set of acceptance BIM model spatial attenuation pixel granularity semantic association score vectors, including: Dividing the acceptance BIM model pixel granularity space span value by a preset scaling factor to obtain the acceptance BIM model pixel granularity space span scaling value; Calculate an exponential function value with e as base and the inverse of the scaling value of the pixel granularity space span of the acceptance BIM model as exponent to obtain a semantic association space attenuation coefficient; The semantic association score vector between pixel granularities of the acceptance BIM model is calculated and divided by the semantic association spatial attenuation coefficient to obtain the semantic association score vector between spatial attenuation pixel granularities of the acceptance BIM model.

7. The data processing method for housing inspection according to claim 6, characterized in that: The set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model is dimensionally reconstructed and weighted to obtain a semantic association weight map between spatial attenuation pixel granularities of the acceptance BIM model structural features, including: After arranging the set of semantic association score vectors between spatial attenuation pixel granularities of the acceptance BIM model into a semantic association score feature map between spatial attenuation pixel granularities of the acceptance BIM model, it is input into a feature dimension modulation layer based on a point convolution layer and a weight allocation network based on a Sigmoid function to obtain a semantic association weight map between spatial attenuation pixel granularities of the acceptance BIM model structural feature.

8. The data processing method for housing inspection according to claim 7, characterized in that: The method of performing feature modulation on the acceptance BIM model structure semantic coding feature map based on the semantic association weight map between attenuated pixel granularities of the acceptance BIM model structure feature space to obtain the acceptance BIM model structure semantic space enhanced coding feature map includes: The acceptance BIM model structure feature space attenuation pixel granularity semantic association weight map and the acceptance BIM model structure semantic coding feature map are calculated by multiplying the points by position to obtain the acceptance BIM model structure semantic space enhanced coding feature map.

9. The data processing method for housing inspection according to claim 8, characterized in that: Determining whether the house to be tested meets the acceptance criteria of the construction party based on the feature difference between the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map includes: Inputting the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map into a BIM model difference measurement network to obtain a difference semantic coding feature map of the acceptance-design house mark; The difference semantic coding feature map of the acceptance-design house mark is input into the classifier-based intelligent auxiliary acceptance module to obtain an auxiliary acceptance result, and the auxiliary acceptance result is used to indicate whether it meets the acceptance standard of the construction party.

10. A data processing system for house acceptance, characterized in that: include: A BIM model acquisition module is used to acquire the acceptance BIM model of the house subject to be tested, and at the same time extract the design BIM model of the house subject to be tested from the database; A housing structure semantic feature extraction module is used to extract the housing structure semantic features of the acceptance BIM model and the design BIM model to obtain a structural semantic coding feature graph of the acceptance BIM model and a structural semantic coding feature graph of the design BIM model; An attention enhancement module is used to perform attention enhancement on the acceptance BIM model structure semantic coding feature map and the design BIM model structure semantic coding feature map to obtain the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map; The acceptance standard evaluation module is used to determine whether the house subject to be tested meets the acceptance standard of the construction party based on the feature difference between the acceptance BIM model structure semantic space enhanced coding feature map and the design BIM model structure semantic space enhanced coding feature map.

Images