Method for repairing defective part of building component

Abstract

The present invention discloses a method for repairing defective parts of building components, comprising collecting images of defective building components from multiple angles; creating a three-dimensional model of the defective building component and extracting its contours; identifying the finish type of the defective building component using multiple models; using a trained deep learning model to search a component model database for a component model that is closest in style and finish to the defective building component based on the finish type and contours, as a repair example; repairing the defective part in the three-dimensional model of the defective building component according to the style of the repair example to form a three-dimensional image of the defective part; printing and producing the defective part using 3D printing technology; installing the defective part at the defective part of the building component; and repairing the outer surface of the defective part with a finishing material according to the finish type of the repair example. The present invention can improve the repair quality of the style and finish type of the building component, ensuring the overall shape and decorative effect of the finish.

Description

Technical Field

[0001] The present invention relates to the field of building maintenance, and in particular to a method for repairing a defective part of a building component. Background Art

[0002] As building structures age, structural and finishing defects may occur in various components. Currently, the inspection of the structural shape and finishing type of defective building components is performed manually, resulting in low detection efficiency. The judgment of the structural style and finishing material of the defective parts relies on construction experience, which often leads to misjudgments. Repair work on the defective parts based on these misjudgments results in poor matching of the structural style and finishing type of the repaired parts with the original building components, resulting in errors and affecting the overall shape and finishing of the existing building. Repairs to defective parts of existing buildings are typically performed by filling, which can result in quality defects such as cracks between the filled areas and the original building, affecting the decorative effect of the existing building's finishes. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for repairing defective parts of building components to solve the problem that manual inspection of the structural modeling and finishing type of defective parts of building components is inefficient and has large errors, which affects the overall structural modeling and finishing decoration of the defective parts of building components after repair.

[0004] In order to solve the above technical problems, the present invention provides a method for repairing a defective portion of a building component, comprising:

[0005] Capture images of defective building components containing defective parts from multiple angles;

[0006] Establishing a three-dimensional model of the defective building components based on images of the defective building components collected from multiple angles;

[0007] Extract the contour lines of the 3D model of the defective building components;

[0008] The collected images of defective building components are used to identify the finishing types of the defective building components through the trained multi-model;

[0009] Based on the finish type and the outline of the 3D model of the defective building component, the trained deep learning model is used to search the component model database for the model that is closest to the style and finish type of the defective building component as a repair example;

[0010] According to the modeling style of the repair sample, the defective parts in the three-dimensional model of the defective building components are repaired to form an independent three-dimensional image of the defective parts;

[0011] Use 3D printing technology to print the three-dimensional image of the defective part to produce the defective part;

[0012] Installing the defective component at the defective portion of the building component;

[0013] Repair the outer surface of the damaged part with finishing materials according to the finishing type of the repair sample.

[0014] Furthermore, the method for repairing defective parts of building components provided by the present invention uses a drone equipped with a high-precision camera to capture images of the defective building components containing the defective parts from multiple angles.

[0015] Furthermore, the method for repairing defective parts of building components provided by the present invention includes establishing a three-dimensional model diagram of the defective building components based on images of the defective building components collected from multiple angles, which includes: analyzing and processing the collected images of the defective building components through the SFM algorithm to establish a three-dimensional model diagram of the defective building components.

[0016] Furthermore, the method for repairing a defective portion of a building component provided by the present invention includes: using a trained multi-model to identify the type of finish of the defective building component from the collected image of the defective building component;

[0017] The images of defective building components are collected and the types of finishes of the defective building components are determined by the trained multi-model.

[0018] When the judgment results of the models on the finishing type are the same, the judgment result is determined to be the finishing type of the building component;

[0019] When the judgment results of each model are different, according to the probability fusion formula (1), the one with the highest probability among the judgment results is taken as the finishing type of the building component;

[0020] (C i )= (1)

[0021] In formula (1), C i Is the finish type, (C i ) is used to judge the image of the building component with defective parts as C i Probability of finish type, is the kth model for the facing category The probability output of is the weight of the k-th model.

[0022] Furthermore, in the method for repairing defective parts of building components provided by the present invention, the trained multi-model includes a main model and an auxiliary model, the main model is YOLOv8, and the auxiliary model is EfficientNet-B4.

[0023] Furthermore, the method for repairing a defective portion of a building component provided by the present invention includes extracting the contour line of a three-dimensional model of the defective building component, comprising:

[0024] Convert the three-dimensional model of the defective building component into a two-dimensional image, and detect the edge contour of the two-dimensional image using an edge detection algorithm;

[0025] Perform morphological operations on the edge contour detection results to remove noise and broken edges and connect discontinuous contour lines;

[0026] The contours of the missing building components are extracted and the modeling style of their shapes is identified through Fourier descriptors.

[0027] Compared with the prior art, the method for repairing defective parts of building components provided by the present invention has the following beneficial effects:

[0028] The multi-model is used to identify the finishing type of the defective building component from the collected images of the defective building component, thereby improving the accuracy of the finishing type detection of the defective building component and reducing the judgment error of the finishing type.

[0029] By using the contour lines and finishing types of the three-dimensional model of the defective building component as input, the trained deep learning model is used to search the component model database for the modeling style and finishing type that are closest to the contour line shape of the defective building component as a repair sample, thereby determining the modeling style of the defective building component, improving the detection efficiency and accuracy of the modeling style of the defective building component, and reducing the errors of manual detection and judgment.

[0030] The defective parts in the three-dimensional model of the defective building component are completed and repaired by using the modeling style of the repair sample to form a three-dimensional map of the defective part that can be removed from the three-dimensional model of the defective building component. The three-dimensional map of the defective part is printed by 3D printing technology to produce the defective parts, thereby improving the production quality and production accuracy of the defective parts of the building component. It is installed on the building component to make the two perfectly match, thereby improving the repair quality and repair accuracy of the defective part and ensuring the overall modeling style of the repaired building component.

[0031] The finishing material of the outer surface of the defective part is repaired by using the finishing type of the repair sample, so that the repaired finishing material is consistent with the original finishing material of the defective building component or as close as possible to the original finishing material, thereby ensuring the overall finishing effect of the repaired building component. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a flow chart of the method for repairing defective parts of building components. DETAILED DESCRIPTION

[0033] The present invention will be described in detail below with reference to the accompanying drawings. The advantages and features of the present invention will become more apparent from the following description. It should be noted that the drawings are all in a very simplified form and are not accurately scaled, and are only used to facilitate and clearly illustrate the embodiments of the present invention.

[0034] Please refer to Figure 1 The embodiment of the present invention provides a method for repairing a defective portion of a building component, which may include:

[0035] Step S1: Capture images of the defective building component with the defective part from multiple angles. To improve acquisition accuracy and convenience, a high-precision camera mounted on a drone can be used to capture images of the defective building component with the defective part from multiple angles.

[0036] Step S2: Create a 3D model of the defective building component based on the images of the defective building component captured from multiple angles. This may include analyzing and processing the captured images of the defective building component using the SFM algorithm to create the 3D model of the defective building component. SFM (Structure from Motion) is a computer vision technology that recovers the 3D geometric structure of a scene and the camera's motion trajectory from a multi-view 2D image sequence. This technology analyzes feature matching relationships and geometric constraints between images to jointly optimize the camera's position and posture and the 3D point coordinates of the scene, achieving 3D modeling from 2D to 3D.

[0037] Step S3 extracts the contour lines of the 3D model of the defective building component. Specifically, this may include converting the 3D model of the defective building component into a 2D image and detecting the edge contours of the 2D image using an edge detection algorithm, such as the Canny edge detection algorithm. To improve the accuracy and completeness of the contour lines, morphological operations (dilation and erosion) may be performed on the edge contour detection results to remove noise and broken edges and connect discontinuous contour lines. Finally, the contour lines of the defective building component are extracted using Fourier descriptors. This allows the style of the defective building component to be identified through the contour lines.

[0038] Step S4: The images of defective building components are collected and the type of finish of the defective building components is identified by the trained multi-model. The finish types include but are not limited to water-brushed stone, terrazzo, paint, gold foil finish, etc. The trained multi-model can use YOLOv8 as the main model and EfficientNet-B4 as the auxiliary model. When training the two models, labeled samples of building finishes such as water-brushed stone, terrazzo, paint, and gold foil finish are used for training, focusing on optimizing the detection ability of small targets (such as gold foil finish details). This improves the accuracy of the judgment results of each model. The auxiliary model can output a global category probability (such as 0.95 for gold foil finish and 0.03 for paint) to supplement the global finish type information after the main model identifies the finish type of the defective building component. When the judgment results of the models on the finish type are the same, the judgment result is determined to be the finish type of the building component; when the judgment results of the models are different, according to the probability fusion formula (1), the one with the higher probability in the judgment result is taken as the finish type of the building component.

[0039] (C i )= (1);

[0040] In formula (1), C i Is the finish type, (C i ) is used to judge the image of the building component with defective parts as C i Probability of finish type, is the kth model for the facing category The probability output of is the weight of the kth model. The weights are assigned based on the accuracy of each model on the validation set. For example, the weight of YOLOv8 can be 0.6, and the weight of EfficientNet-B4 can be 0.4.

[0041] If all models are for a certain finish type C i Output probability ( >0.8), then the output image’s finishing material judgment result C i If there is a difference (such as gold leaf finish 0.6 vs. paint 0.5), the maximum value is taken according to the weighted probability. If the weighted probabilities of multiple finishes are close (the probability difference is less than 0.01), manual intervention is required to determine the finish type C i .

[0042] To improve recognition accuracy, the captured image can be enhanced before the multi-model is used to identify the type of finish. For example, contrast-limited adaptive histogram equalization (CLAHE) can be used to enhance the contrast and clarity of the image and highlight the texture and color characteristics of the finish.

[0043] The order of steps S2, S3 and S4 can be interchanged.

[0044] Step S5, based on the finish type and the contour line of the three-dimensional model of the defective building component, the trained deep learning model is used to search the component model database for the component model that is closest to the style and finish type of the defective building component as a repair sample. In the component model database, a large number of component models with different finish types and styles and their corresponding feature vectors are stored, including the finish material type, length, width, perimeter, symmetry, regularity, curvature change rate, and the size of the area enclosed by the contour line. The similarity between the feature vector of the component to be repaired and the feature vector of each component model in the database is calculated by cosine similarity, and the component model with the highest similarity is found as the component model of the similar type as the repair sample. The trained deep learning model can be BERT.

[0045] In step S6, the defective parts of the 3D model of the defective building component are repaired according to the modeling style of the repair example to form a 3D image of the defective part. Materials with appropriate strength, durability, and appearance characteristics can be selected based on the actual use environment and requirements of the defective building component. For example, for components that bear heavy loads, high-strength plastic or metal materials can be selected; for components with high aesthetic requirements, resin materials with good surface quality can be selected.

[0046] Step S7: 3D printing the defective part using the three-dimensional image of the defective part. When designing the 3D printed model, the connection surface is designed to be slightly larger than the three-dimensional image of the defective part, generally extending 10-20 mm beyond the edge of the defective part. This ensures better connection and fixation with the original component during installation.

[0047] Step S8, install the defective part at the defective part of the building component. The specific installation method may include: selecting a suitable connection method according to whether the angle of the defective part is close to the vertical plane or the horizontal plane. If the defective part is close to the vertical plane, a bolt + glue connection method can be used. First, drill holes on the original component and the defective part produced by 3D printing respectively, and then use bolts to preliminarily fix the two, and then apply high-strength glue on the connection part to enhance the stability and sealing of the connection. If the defective part is on the horizontal plane, a hidden step + bolt connection method can be used. A hidden step is machined on the original component, and the corresponding part of the defective part produced by 3D printing is embedded in the step, and then fixed with bolts. This connection method can improve the strength and stability of the connection, while making the connection part more concealed and not affecting the appearance of the component.

[0048] Step S9: Repairing the outer surface of the defective component with finishing materials according to the finishing type of the repair sample.

[0049] The method for repairing defective parts of building components provided in an embodiment of the present invention uses multiple models to identify the finishing type of the defective building components from the collected images of the defective building components, thereby improving the accuracy of detecting the finishing type of the defective building components and reducing the judgment error of the finishing type.

[0050] The method for repairing defective parts of building components provided by an embodiment of the present invention uses the contour lines and finish types of the three-dimensional model of the defective building component as input, and uses a trained deep learning model to search a component model database for a component model that is closest in style and finish type to the contour line shape of the defective building component as a repair sample, thereby determining the style of the defective building component, improving the detection efficiency and accuracy of the style of the defective building component, and reducing errors in manual detection and judgment.

[0051] The method for repairing defective parts of building components provided by an embodiment of the present invention completes and repairs the defective parts in a three-dimensional model diagram of a defective building component through the modeling style of a repair sample to form a three-dimensional diagram of the defective part that can be removed from the three-dimensional model diagram of the defective building component. The three-dimensional diagram of the defective part is printed using 3D printing technology to produce defective parts, thereby improving the production quality and production accuracy of the defective part of the building component. The defective part is installed on the building component so that the two are perfectly matched, thereby improving the repair quality and repair accuracy of the defective part and ensuring the overall modeling style of the repaired building component.

[0052] The method for repairing defective parts of building components provided in an embodiment of the present invention repairs the outer surface of the defective part with finishing materials by repairing the finishing type of the repair sample, so that the repaired finishing materials are consistent with the original finishing materials of the defective building component or are as close as possible to the original finishing materials, thereby ensuring the overall finishing and decorative effect of the repaired building component.

[0053] The method for repairing defective parts of building components provided by the embodiment of the present invention can realize the full process intelligence and precision of repairing defective building components, effectively improve the efficiency and quality of repair, and ensure that the repaired components are highly consistent with the original components in terms of structural strength and appearance style.

[0054] The present invention is not limited to the specific embodiments described above. Obviously, the embodiments described above are only some embodiments of the embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention described, all other embodiments obtained by ordinary technicians in this field fall within the scope of protection of the present invention. Those skilled in the art can make other levels of modifications and changes to the present invention. In this way, if these modifications and changes of the present invention fall within the scope of the claims of the present invention, the present invention is also intended to include these changes and changes.

Claims

1. A method for repairing a defective part of a building component, characterized in that: include: Capture images of defective building components containing defective parts from multiple angles; Establishing a three-dimensional model of the defective building components based on images of the defective building components collected from multiple angles; Extract the contour lines of the 3D model of the defective building components; The images of defective building components are collected and the types of finishes of the defective building components are determined by the trained multi-model. When the judgment results of the models on the finishing type are the same, the judgment result is determined to be the finishing type of the building component; When the judgment results of each model are different, according to the probability fusion formula, the one with the highest probability is taken as the finishing type of the building component; Based on the finish type and the outline of the 3D model of the defective building component, the trained deep learning model is used to search the component model database for the model that is closest to the style and finish type of the defective building component as a repair example; According to the modeling style of the repair sample, the defective parts in the three-dimensional model of the defective building components are repaired to form an independent three-dimensional image of the defective parts; Use 3D printing technology to print the three-dimensional image of the defective part to produce the defective part; Installing the defective component at the defective portion of the building component; Repair the outer surface of the damaged part with finishing materials according to the finishing type of the repair sample.

2. The method for repairing a defective portion of a building component according to claim 1, wherein: Use a drone equipped with a high-precision camera to collect images of defective building components containing defective parts from multiple angles.

3. The method for repairing a defective portion of a building component according to claim 1, wherein: The method for establishing a three-dimensional model of a defective building component based on images of the defective building component collected from multiple angles includes: analyzing and processing the collected images of the defective building component by an SFM algorithm to establish a three-dimensional model of the defective building component.

4. The method for repairing a defective portion of a building component according to claim 1, wherein: The probability fusion formula (1) is: (C i )= （1） In formula (1), C i Is the finish type, (C i ) is used to judge the image of the building component with defective parts as C i Probability of finish type, is the kth model for the facing category The probability output of is the weight of the k-th model.

5. The method for repairing a defective portion of a building component according to claim 1, wherein: The trained multi-model includes a main model and an auxiliary model, the main model is YOLOv8, and the auxiliary model is EfficientNet-B4.

6. The method for repairing a defective portion of a building component according to claim 1, wherein: The method for extracting the contour line of the three-dimensional model of the defective building component includes: Convert the three-dimensional model of the defective building component into a two-dimensional image, and detect the edge contour of the two-dimensional image using an edge detection algorithm; Perform morphological operations on the edge contour detection results to remove noise and broken edges and connect discontinuous contour lines; The contours of the missing building components are extracted and the modeling style of their shapes is identified through Fourier descriptors.

Images